From 2dc0a04d9bf45f5e8b434e55896555b3341255a6 Mon Sep 17 00:00:00 2001 From: Bhaskar Ray Date: Fri, 26 Jun 2026 17:32:53 -0400 Subject: [PATCH] add grouped exploration of bivariate data, edit optional chunk in EDA Solutions --- .../exploratory-data-analysis-solutions.ipynb | 4283 +++++++++-------- 1 file changed, 2260 insertions(+), 2023 deletions(-) diff --git a/workshops/exploratory-data-analysis/exploratory-data-analysis-solutions.ipynb b/workshops/exploratory-data-analysis/exploratory-data-analysis-solutions.ipynb index 4980288..2b12b0d 100644 --- a/workshops/exploratory-data-analysis/exploratory-data-analysis-solutions.ipynb +++ b/workshops/exploratory-data-analysis/exploratory-data-analysis-solutions.ipynb @@ -1,2055 +1,2292 @@ { - "cells": [ - { - "cell_type": "markdown", - "id": "6f248326", - "metadata": {}, - "source": [ - "# Getting Started with Python: Exploratory Data Analysis (Polars)\n", - "\n", - "**Author:** Alp Tezbasaran \n", - "**Last updated:** 2026-02-10 \n", - "\n", - "This notebook is designed for a **live, in-person workshop**. It is intentionally iterative: we’ll ask a question, inspect the data, try a quick transformation or plot, then repeat.\n" - ] - }, - { - "cell_type": "markdown", - "id": "c5cb57c1", - "metadata": {}, - "source": [ - "## 0. Accessibility: theme, text size, and output\n", - "\n", - "- **Tools → Settings → Theme**: Light / Dark / System\n", - "- **Tools → Settings → Editor → Font size**: increase if needed\n", - "- **Browser zoom**: Cmd/Ctrl `+` or `-`\n", - "\n", - "Teaching tip: agree on a standard zoom level at the start.\n" - ] - }, - { - "cell_type": "markdown", - "id": "b3dafe05", - "metadata": {}, - "source": [ - "## Learning outcomes\n", - "\n", - "By the end of this workshop, you will be able to:\n", - "\n", - "- Load tabular data into Polars\n", - "- Profile data (shape, schema, missingness, unique values)\n", - "- Use **expressions** to filter and create features\n", - "- Summarize with `group_by().agg()` and explain what information is lost\n", - "- Use pivoting as **“groupby + aggregation + reshaping”**\n", - "- Make quick univariate and bivariate plots for exploration\n", - "- Recognize when to use **lazy** execution for performance\n" - ] - }, - { - "cell_type": "markdown", - "id": "25e87d09", - "metadata": {}, - "source": [ - "## The EDA loop (a simple roadmap)\n", - "\n", - "EDA is not linear. A useful loop is:\n", - "\n", - "1. **Inventory**: What columns do we have? What types? How big?\n", - "2. **Quality**: Missing values, duplicates, weird values\n", - "3. **Univariate**: What does each variable look like by itself?\n", - "4. **Bivariate / multivariate**: What relationships appear?\n", - "5. **Feature ideas**: Create features that clarify relationships\n", - "6. **Summarize & communicate**: tables/plots + short written takeaways\n", - "\n", - "```mermaid\n", - "flowchart LR\n", - " A[Inventory] --> B[Quality]\n", - " B --> C[Univariate]\n", - " C --> D[Bivariate / multivariate]\n", - " D --> E[Feature ideas]\n", - " E --> F[Summarize & communicate]\n", - " F --> A\n", - "```\n", - "\n", - "We’ll follow that loop on a few small datasets.\n" - ] - }, - { - "cell_type": "markdown", - "id": "b84b21f5", - "metadata": {}, - "source": [ - "## 1. Setup\n", - "\n", - "We’ll use **Polars** for data work and **matplotlib** for plotting.\n", - "\n", - "If you’re in Google Colab, install Polars once per session.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 35, - "id": "9e1c1683", - "metadata": {}, - "outputs": [], - "source": [ - "import polars as pl\n", - "import matplotlib.pyplot as plt\n", - "from pathlib import Path\n", - "\n", - "data_dir = Path('data')\n", - "\n", - "plt.rcParams['figure.dpi'] = 120\n", - "plt.rcParams['axes.grid'] = True" - ] - }, - { - "cell_type": "markdown", - "id": "f33be526", - "metadata": {}, - "source": [ - "## Polars: the 3 practical pillars\n", - "\n", - "Think of Polars as: **operation + expression + execution mode**.\n", - "\n", - "This gives students a simple mental model:\n", - "\n", - "1. **Operation (what you do)**\n", - " - DataFrame/LazyFrame methods define the step: `select`, `filter`, `with_columns`, `group_by`, `join`.\n", - "2. **Expression (how you describe it)**\n", - " - Expressions define the rule/math: `pl.col(...)`, `pl.when(...).then(...).otherwise(...)`, boolean logic, arithmetic.\n", - "3. **Execution mode (when/how it runs)**\n", - " - Eager runs now.\n", - " - Lazy builds a plan and runs at `collect()`, which helps Polars optimize.\n", - "\n", - "```mermaid\n", - "flowchart LR\n", - " A[Operation: select/filter/group_by] --> B[Expression: pl.col + rules]\n", - " B --> C[Execution mode: eager or lazy]\n", - "```\n", - "\n", - "### Easy analogy: cooking\n", - "\n", - "- **Operation** = the cooking action (chop, mix, bake)\n", - "- **Expression** = the recipe rule (\"add 2 cups\", \"if too salty, add water\")\n", - "- **Execution mode** = when you cook\n", - " - eager: cook each step immediately\n", - " - lazy: plan the full recipe first, then cook efficiently\n", - "\n", - "For today, we focus on the first two pillars most of the time.\n", - "We introduce lazy later as a performance bonus, not a required concept.\n" - ] - }, - { - "cell_type": "markdown", - "id": "1d2a27ac", - "metadata": {}, - "source": [ - "### Expression primer (5 minutes)\n", - "\n", - "In Polars, you rarely write row-by-row loops. You write **column expressions**.\n", - "\n", - "Common building blocks:\n", - "- `pl.col('x')` → refer to a column\n", - "- `.alias('new_name')` → rename an expression result\n", - "- Boolean logic: `&` (and), `|` (or), `~` (not)\n", - "- **Use parentheses** with boolean logic: `(cond1) & (cond2)`\n", - "- Conditional: `pl.when(cond).then(val).otherwise(val)`\n", - "- String/date helpers: `.str.*`, `.dt.*`\n" - ] - }, - { - "cell_type": "code", - "execution_count": 36, - "id": "2f603b66", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (4, 5)
xsx_filleds_cleanflag
i64stri64strstr
1" A "1"a""ok"
2"b"2"b""ok"
3"b"3"b""ok"
nullnull0"missing""was_null"
" - ], - "text/plain": [ - "shape: (4, 5)\n", - "┌──────┬──────┬──────────┬─────────┬──────────┐\n", - "│ x ┆ s ┆ x_filled ┆ s_clean ┆ flag │\n", - "│ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", - "│ i64 ┆ str ┆ i64 ┆ str ┆ str │\n", - "╞══════╪══════╪══════════╪═════════╪══════════╡\n", - "│ 1 ┆ A ┆ 1 ┆ a ┆ ok │\n", - "│ 2 ┆ b ┆ 2 ┆ b ┆ ok │\n", - "│ 3 ┆ b ┆ 3 ┆ b ┆ ok │\n", - "│ null ┆ null ┆ 0 ┆ missing ┆ was_null │\n", - "└──────┴──────┴──────────┴─────────┴──────────┘" - ] - }, - "execution_count": 36, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "expr_demo = pl.DataFrame({'x':[1,2,3,None], 's':[' A ','b','b',None]})\n", - "(\n", - " expr_demo\n", - " .with_columns([\n", - " pl.col('x').fill_null(0).alias('x_filled'),\n", - " pl.col('s').str.to_lowercase().str.strip_chars().fill_null('missing').alias('s_clean'),\n", - " pl.when(pl.col('x').is_null()).then(pl.lit('was_null')).otherwise(pl.lit('ok')).alias('flag')\n", - " ])\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "969f55dd", - "metadata": {}, - "source": [ - "## Checkpoint A (quick)\n", - "\n", - "- Everyone can run the setup cell and the expression demo\n", - "- If you’re stuck on imports/paths, follow along and we’ll fix it during the next pause\n" - ] - }, - { - "cell_type": "markdown", - "id": "f2839e6f", - "metadata": {}, - "source": [ - "## 2. Load datasets\n", - "\n", - "We’ll use three small datasets:\n", - "\n", - "- `NCSU_Mascots_v1.csv` *(synthetic)*\n", - "- `NCSU Celebrity Graduates_v1.csv` *(synthetic)*\n", - "- `penguins.csv` *(public dataset, light cleaning demo)*\n", - "\n", - "Even with synthetic data, practice good habits: avoid re-identification and think about bias.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 37, - "id": "0e4e3911", - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "((50, 22), (100, 16), (344, 7))" - ] - }, - "execution_count": 37, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "mascots = pl.read_csv(data_dir / 'NCSU_Mascots_v1.csv')\n", - "celebs = pl.read_csv(data_dir / 'NCSU Celebrity Graduates_v1.csv')\n", - "penguins_raw = pl.read_csv(data_dir / 'penguins.csv')\n", - "\n", - "(mascots.shape, celebs.shape, penguins_raw.shape)" - ] - }, - { - "cell_type": "markdown", - "id": "3b93421f", - "metadata": {}, - "source": [ - "## 3. Inventory & profiling\n", - "\n", - "Start every EDA with:\n", - "- **shape** (rows, columns)\n", - "- **schema** (column types)\n", - "- **preview** (head/tail)\n", - "\n", - "Then a quick profile:\n", - "- missingness (counts *and* rates)\n", - "- numeric summaries\n", - "- categorical cardinality (how many unique values?)\n", - "- duplicates (do we expect any?)\n" - ] - }, - { - "cell_type": "code", - "execution_count": 38, - "id": "773bf757", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Mascots: (50, 22)\n", - "Schema({'Name': String, 'Species': String, 'Breed': String, 'Prior Position': String, 'Prior Employer': String, 'Interview Date': String, 'Hire Year': Int64, 'Hire Date': String, 'Termination Year': Float64, 'Termination Date': String, 'Termination Reason': String, 'Post Employment Position': String, 'Nick Name': String, 'Temperment': String, 'Fur Color': String, 'Age': Float64, 'Height': Float64, 'Weight': String, 'Offer Sent': Int64, 'Starting Salary': Int64, 'Biter': Int64, 'Hired': Int64})\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "shape: (5, 22)
NameSpeciesBreedPrior PositionPrior EmployerInterview DateHire YearHire DateTermination YearTermination DateTermination ReasonPost Employment PositionNick NameTempermentFur ColorAgeHeightWeightOffer SentStarting SalaryBiterHired
strstrstrstrstrstri64strf64strstrstrstrstrstrf64f64stri64i64i64i64
"Wallace Whitfield Riddick""Human"nullnullnull"1899-02-17"1899"1899-02-23"nullnullnullnullnull"patient""cream"5.034.5"40.0"1948101
"Tige""Dog""English Bulldog"nullnull"1910-02-16"1910"1910-02-17"nullnullnullnullnull"sweet""brown"10.914.2"51.0"11013001
"Togo""Dog""American Bulldog"nullnull"1910-06-28"1910"1910-07-01"nullnullnullnullnull"sweet""brown"10.424.4"90.0"1957201
"State""Wolf""Timber Wolf"nullnull"1940-05-02"1940"1940-05-03"1946.0"1940-05-17""New Job""Traveling Zoo"null"bewildered""multi-color"9.829.8"112.5"1992101
"Lobo I""Wolf""Timber Wolf""Zoo""Philadelphia Zoo""1959-08-26"1959"1959-09-03"1959.0"1959-05-29""Passed Away"nullnull"quiet""gray"10.327.2"100.15"1950901
" - ], - "text/plain": [ - "shape: (5, 22)\n", - "┌───────────┬─────────┬──────────────────┬──────────┬───┬────────────┬──────────┬───────┬───────┐\n", - "│ Name ┆ Species ┆ Breed ┆ Prior ┆ … ┆ Offer Sent ┆ Starting ┆ Biter ┆ Hired │\n", - "│ --- ┆ --- ┆ --- ┆ Position ┆ ┆ --- ┆ Salary ┆ --- ┆ --- │\n", - "│ str ┆ str ┆ str ┆ --- ┆ ┆ i64 ┆ --- ┆ i64 ┆ i64 │\n", - "│ ┆ ┆ ┆ str ┆ ┆ ┆ i64 ┆ ┆ │\n", - "╞═══════════╪═════════╪══════════════════╪══════════╪═══╪════════════╪══════════╪═══════╪═══════╡\n", - "│ Wallace ┆ Human ┆ null ┆ null ┆ … ┆ 1 ┆ 9481 ┆ 0 ┆ 1 │\n", - "│ Whitfield ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │\n", - "│ Riddick ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │\n", - "│ Tige ┆ Dog ┆ English Bulldog ┆ null ┆ … ┆ 1 ┆ 10130 ┆ 0 ┆ 1 │\n", - "│ Togo ┆ Dog ┆ American Bulldog ┆ null ┆ … ┆ 1 ┆ 9572 ┆ 0 ┆ 1 │\n", - "│ State ┆ Wolf ┆ Timber Wolf ┆ null ┆ … ┆ 1 ┆ 9921 ┆ 0 ┆ 1 │\n", - "│ Lobo I ┆ Wolf ┆ Timber Wolf ┆ Zoo ┆ … ┆ 1 ┆ 9509 ┆ 0 ┆ 1 │\n", - "└───────────┴─────────┴──────────────────┴──────────┴───┴────────────┴──────────┴───────┴───────┘" - ] - }, - "execution_count": 38, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "print('Mascots:', mascots.shape)\n", - "print(mascots.schema)\n", - "mascots.head(5)" - ] - }, - { - "cell_type": "code", - "execution_count": 39, - "id": "279d69ce", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (5, 22)
NameSpeciesBreedPrior PositionPrior EmployerInterview DateHire YearHire DateTermination YearTermination DateTermination ReasonPost Employment PositionNick NameTempermentFur ColorAgeHeightWeightOffer SentStarting SalaryBiterHired
strstrstrstrstrstri64strf64strstrstrstrstrstrf64f64stri64i64i64i64
"Shadow""Spider""Golden Orb Weaver""Web Infrastructure Designer""Giggle ""2013-04-12"nullnullnullnullnullnull"Gizmo""devilishly clever"null11.44.0"0.009"0null00
"Clover""Rabbit""Mini Rex""Burrow Engineer""Hillside Civil Engineering""1952-07-30"nullnullnullnullnullnullnull"frightened""brown"10.012.4"3.71"0null00
"Sophie""Rabbit""Holland Lop""Carrot Quality Control Associa…"Publix""1953-11-17"nullnullnullnullnullnull"Spock""social""black"9.710.55"3.18"1990400
"Pixel""Turtle""Eastern Box""Rock Impersonator""No Moves, LLC""1997-08-25"nullnullnullnullnullnull"Gizmo""patient"null11.65.44"1.49"0null00
"Baron""Dog""Dalmatian""Firehouse Morale Officer""Raleigh Fire Department""1990-08-06"nullnullnullnullnullnullnull"social""black"14.721.5"57.68"0null00
" - ], - "text/plain": [ - "shape: (5, 22)\n", - "┌────────┬─────────┬─────────────┬─────────────────────┬───┬────────────┬──────────┬───────┬───────┐\n", - "│ Name ┆ Species ┆ Breed ┆ Prior Position ┆ … ┆ Offer Sent ┆ Starting ┆ Biter ┆ Hired │\n", - "│ --- ┆ --- ┆ --- ┆ --- ┆ ┆ --- ┆ Salary ┆ --- ┆ --- │\n", - "│ str ┆ str ┆ str ┆ str ┆ ┆ i64 ┆ --- ┆ i64 ┆ i64 │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ i64 ┆ ┆ │\n", - "╞════════╪═════════╪═════════════╪═════════════════════╪═══╪════════════╪══════════╪═══════╪═══════╡\n", - "│ Shadow ┆ Spider ┆ Golden Orb ┆ Web Infrastructure ┆ … ┆ 0 ┆ null ┆ 0 ┆ 0 │\n", - "│ ┆ ┆ Weaver ┆ Designer ┆ ┆ ┆ ┆ ┆ │\n", - "│ Clover ┆ Rabbit ┆ Mini Rex ┆ Burrow Engineer ┆ … ┆ 0 ┆ null ┆ 0 ┆ 0 │\n", - "│ Sophie ┆ Rabbit ┆ Holland Lop ┆ Carrot Quality ┆ … ┆ 1 ┆ 9904 ┆ 0 ┆ 0 │\n", - "│ ┆ ┆ ┆ Control Associa… ┆ ┆ ┆ ┆ ┆ │\n", - "│ Pixel ┆ Turtle ┆ Eastern Box ┆ Rock Impersonator ┆ … ┆ 0 ┆ null ┆ 0 ┆ 0 │\n", - "│ Baron ┆ Dog ┆ Dalmatian ┆ Firehouse Morale ┆ … ┆ 0 ┆ null ┆ 0 ┆ 0 │\n", - "│ ┆ ┆ ┆ Officer ┆ ┆ ┆ ┆ ┆ │\n", - "└────────┴─────────┴─────────────┴─────────────────────┴───┴────────────┴──────────┴───────┴───────┘" - ] - }, - "execution_count": 39, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "mascots.tail(5)" - ] - }, - { - "cell_type": "code", - "execution_count": 40, - "id": "708a7be5", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Celebs: (100, 16)\n", - "Schema({'#': Int64, 'Name': String, 'Birth Year': Int64, 'Death Year': Float64, 'Occupation(s)': String, 'Country Born In': String, 'Birthdate': String, 'Age': Int64, 'Graduation Month/Year': String, 'Workstudy Position': String, 'Workstudy Hourly Rate': Float64, 'GPA': Float64, 'College': String, 'Loved Library?': Int64, 'Favorite NCSU Hangout': String, 'Favorite Raleigh Hangout': String})\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "shape: (5, 16)
#NameBirth YearDeath YearOccupation(s)Country Born InBirthdateAgeGraduation Month/YearWorkstudy PositionWorkstudy Hourly RateGPACollegeLoved Library?Favorite NCSU HangoutFavorite Raleigh Hangout
i64stri64f64strstrstri64strstrf64f64stri64strstr
88"Adrian Wilson"1979null"American football player""United States""1979-10-12"45"Dec/1997""IT"15.523.88"Humanities and Social Sciences"0"Library""Cup A Joe"
95"Ahmet Özal"1955null"businessperson, politician""Turkey""1955-09-05"69"Dec/1974""Campus Events"16.443.54"Education"0"Fountain Dining Hall""The Optimist"
72"Akihiro Kitamura"1979null"seiyū, actor, television actor""Japan""1979-03-25"45"Dec/1997""IT"16.892.17"Natural Resources"0"Free Expression Tunnel""The Optimist"
10"Anna Camp"1982null"film actor, television actor, …"United States""1982-09-27"42"Dec/2000""Campus Events"14.842.18"Veterinary Medicine"0"Fountain Dining Hall""Mitch's Tavern"
5"Anthony Mackie"1978null"stage actor, film actor, telev…"United States""1978-09-23"46"Dec/1996""Library"17.793.55"Textiles"0"The Brickyard""State Farmers Market Restauran…
" - ], - "text/plain": [ - "shape: (5, 16)\n", - "┌─────┬─────────────┬───────┬─────────────┬───┬─────────────┬────────────┬────────────┬────────────┐\n", - "│ # ┆ Name ┆ Birth ┆ Death Year ┆ … ┆ College ┆ Loved ┆ Favorite ┆ Favorite │\n", - "│ --- ┆ --- ┆ Year ┆ --- ┆ ┆ --- ┆ Library? ┆ NCSU ┆ Raleigh │\n", - "│ i64 ┆ str ┆ --- ┆ f64 ┆ ┆ str ┆ --- ┆ Hangout ┆ Hangout │\n", - "│ ┆ ┆ i64 ┆ ┆ ┆ ┆ i64 ┆ --- ┆ --- │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ str ┆ str │\n", - "╞═════╪═════════════╪═══════╪═════════════╪═══╪═════════════╪════════════╪════════════╪════════════╡\n", - "│ 88 ┆ Adrian ┆ 1979 ┆ null ┆ … ┆ Humanities ┆ 0 ┆ Library ┆ Cup A Joe │\n", - "│ ┆ Wilson ┆ ┆ ┆ ┆ and Social ┆ ┆ ┆ │\n", - "│ ┆ ┆ ┆ ┆ ┆ Sciences ┆ ┆ ┆ │\n", - "│ 95 ┆ Ahmet Özal ┆ 1955 ┆ null ┆ … ┆ Education ┆ 0 ┆ Fountain ┆ The │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Dining ┆ Optimist │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Hall ┆ │\n", - "│ 72 ┆ Akihiro ┆ 1979 ┆ null ┆ … ┆ Natural ┆ 0 ┆ Free ┆ The │\n", - "│ ┆ Kitamura ┆ ┆ ┆ ┆ Resources ┆ ┆ Expression ┆ Optimist │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Tunnel ┆ │\n", - "│ 10 ┆ Anna Camp ┆ 1982 ┆ null ┆ … ┆ Veterinary ┆ 0 ┆ Fountain ┆ Mitch's │\n", - "│ ┆ ┆ ┆ ┆ ┆ Medicine ┆ ┆ Dining ┆ Tavern │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Hall ┆ │\n", - "│ 5 ┆ Anthony ┆ 1978 ┆ null ┆ … ┆ Textiles ┆ 0 ┆ The ┆ State │\n", - "│ ┆ Mackie ┆ ┆ ┆ ┆ ┆ ┆ Brickyard ┆ Farmers │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Market │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Restauran… │\n", - "└─────┴─────────────┴───────┴─────────────┴───┴─────────────┴────────────┴────────────┴────────────┘" - ] - }, - "execution_count": 40, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "print('Celebs:', celebs.shape)\n", - "print(celebs.schema)\n", - "celebs.head(5)" - ] - }, - { - "cell_type": "code", - "execution_count": 41, - "id": "0b3cc41e", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (5, 16)
#NameBirth YearDeath YearOccupation(s)Country Born InBirthdateAgeGraduation Month/YearWorkstudy PositionWorkstudy Hourly RateGPACollegeLoved Library?Favorite NCSU HangoutFavorite Raleigh Hangout
i64stri64f64strstrstri64strstrf64f64stri64strstr
70"Vinny Del Negro"1966null"basketball player, basketball …"United States""1966-08-09"58"Dec/1984""Dining"15.93.49"Sciences"1"The Brickyard""Jubala Coffee"
69"Vivian Howard"1978null"restaurateur, cook"null"1978-03-18"46"Dec/1996""Dining"17.223.37"Humanities and Social Sciences"0"The Brickyard""Raleigh Beer Garden"
90"Walter B. Jones"19432019.0"politician""United States""1943-02-10"81"Dec/1962""Campus Events"17.433.46"Design"0"Reynolds Coliseum""Irish Coffee Lab"
65"Yasonna Laoly"1953null"politician, political candidat…"Indonesia""1953-05-27"71"Dec/1972""Theater"16.892.57"Natural Resources"0"Free Expression Tunnel""Boxcar"
6"Zach Galifianakis"1969null"voice actor, screen writer, fi…"United States""1969-10-01"55"May/1987""Dining"14.242.42"Design"1"The Brickyard""The Optimist"
" - ], - "text/plain": [ - "shape: (5, 16)\n", - "┌─────┬─────────────┬───────┬─────────────┬───┬─────────────┬────────────┬────────────┬────────────┐\n", - "│ # ┆ Name ┆ Birth ┆ Death Year ┆ … ┆ College ┆ Loved ┆ Favorite ┆ Favorite │\n", - "│ --- ┆ --- ┆ Year ┆ --- ┆ ┆ --- ┆ Library? ┆ NCSU ┆ Raleigh │\n", - "│ i64 ┆ str ┆ --- ┆ f64 ┆ ┆ str ┆ --- ┆ Hangout ┆ Hangout │\n", - "│ ┆ ┆ i64 ┆ ┆ ┆ ┆ i64 ┆ --- ┆ --- │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ str ┆ str │\n", - "╞═════╪═════════════╪═══════╪═════════════╪═══╪═════════════╪════════════╪════════════╪════════════╡\n", - "│ 70 ┆ Vinny Del ┆ 1966 ┆ null ┆ … ┆ Sciences ┆ 1 ┆ The ┆ Jubala │\n", - "│ ┆ Negro ┆ ┆ ┆ ┆ ┆ ┆ Brickyard ┆ Coffee │\n", - "│ 69 ┆ Vivian ┆ 1978 ┆ null ┆ … ┆ Humanities ┆ 0 ┆ The ┆ Raleigh │\n", - "│ ┆ Howard ┆ ┆ ┆ ┆ and Social ┆ ┆ Brickyard ┆ Beer │\n", - "│ ┆ ┆ ┆ ┆ ┆ Sciences ┆ ┆ ┆ Garden │\n", - "│ 90 ┆ Walter B. ┆ 1943 ┆ 2019.0 ┆ … ┆ Design ┆ 0 ┆ Reynolds ┆ Irish │\n", - "│ ┆ Jones ┆ ┆ ┆ ┆ ┆ ┆ Coliseum ┆ Coffee Lab │\n", - "│ 65 ┆ Yasonna ┆ 1953 ┆ null ┆ … ┆ Natural ┆ 0 ┆ Free ┆ Boxcar │\n", - "│ ┆ Laoly ┆ ┆ ┆ ┆ Resources ┆ ┆ Expression ┆ │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Tunnel ┆ │\n", - "│ 6 ┆ Zach Galifi ┆ 1969 ┆ null ┆ … ┆ Design ┆ 1 ┆ The ┆ The │\n", - "│ ┆ anakis ┆ ┆ ┆ ┆ ┆ ┆ Brickyard ┆ Optimist │\n", - "└─────┴─────────────┴───────┴─────────────┴───┴─────────────┴────────────┴────────────┴────────────┘" - ] - }, - "execution_count": 41, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "celebs.tail(5)" - ] - }, - { - "cell_type": "code", - "execution_count": 42, - "id": "ddb32ec6", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Penguins: (344, 7)\n", - "Schema({'species': String, 'island': String, 'bill_length_mm': Float64, 'bill_depth_mm': Float64, 'flipper_length_mm': Int64, 'body_mass_g': Int64, 'sex': String})\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "shape: (5, 7)
speciesislandbill_length_mmbill_depth_mmflipper_length_mmbody_mass_gsex
strstrf64f64i64i64str
"Adelie""Torgersen"39.118.71813750"MALE"
"Adelie""Torgersen"39.517.41863800"FEMALE"
"Adelie""Torgersen"40.318.01953250"FEMALE"
"Adelie""Torgersen"nullnullnullnullnull
"Adelie""Torgersen"36.719.31933450"FEMALE"
" - ], - "text/plain": [ - "shape: (5, 7)\n", - "┌─────────┬───────────┬────────────────┬───────────────┬───────────────────┬─────────────┬────────┐\n", - "│ species ┆ island ┆ bill_length_mm ┆ bill_depth_mm ┆ flipper_length_mm ┆ body_mass_g ┆ sex │\n", - "│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", - "│ str ┆ str ┆ f64 ┆ f64 ┆ i64 ┆ i64 ┆ str │\n", - "╞═════════╪═══════════╪════════════════╪═══════════════╪═══════════════════╪═════════════╪════════╡\n", - "│ Adelie ┆ Torgersen ┆ 39.1 ┆ 18.7 ┆ 181 ┆ 3750 ┆ MALE │\n", - "│ Adelie ┆ Torgersen ┆ 39.5 ┆ 17.4 ┆ 186 ┆ 3800 ┆ FEMALE │\n", - "│ Adelie ┆ Torgersen ┆ 40.3 ┆ 18.0 ┆ 195 ┆ 3250 ┆ FEMALE │\n", - "│ Adelie ┆ Torgersen ┆ null ┆ null ┆ null ┆ null ┆ null │\n", - "│ Adelie ┆ Torgersen ┆ 36.7 ┆ 19.3 ┆ 193 ┆ 3450 ┆ FEMALE │\n", - "└─────────┴───────────┴────────────────┴───────────────┴───────────────────┴─────────────┴────────┘" - ] - }, - "execution_count": 42, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "print('Penguins:', penguins_raw.shape)\n", - "print(penguins_raw.schema)\n", - "penguins_raw.head(5)" - ] - }, - { - "cell_type": "code", - "execution_count": 43, - "id": "7ffb6cd5", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (5, 7)
speciesislandbill_length_mmbill_depth_mmflipper_length_mmbody_mass_gsex
strstrf64f64i64i64str
"Gentoo""Biscoe"nullnullnullnullnull
"Gentoo""Biscoe"46.814.32154850"FEMALE"
"Gentoo""Biscoe"50.415.72225750"MALE"
"Gentoo""Biscoe"45.214.82125200"FEMALE"
"Gentoo""Biscoe"49.916.12135400"MALE"
" - ], - "text/plain": [ - "shape: (5, 7)\n", - "┌─────────┬────────┬────────────────┬───────────────┬───────────────────┬─────────────┬────────┐\n", - "│ species ┆ island ┆ bill_length_mm ┆ bill_depth_mm ┆ flipper_length_mm ┆ body_mass_g ┆ sex │\n", - "│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", - "│ str ┆ str ┆ f64 ┆ f64 ┆ i64 ┆ i64 ┆ str │\n", - "╞═════════╪════════╪════════════════╪═══════════════╪═══════════════════╪═════════════╪════════╡\n", - "│ Gentoo ┆ Biscoe ┆ null ┆ null ┆ null ┆ null ┆ null │\n", - "│ Gentoo ┆ Biscoe ┆ 46.8 ┆ 14.3 ┆ 215 ┆ 4850 ┆ FEMALE │\n", - "│ Gentoo ┆ Biscoe ┆ 50.4 ┆ 15.7 ┆ 222 ┆ 5750 ┆ MALE │\n", - "│ Gentoo ┆ Biscoe ┆ 45.2 ┆ 14.8 ┆ 212 ┆ 5200 ┆ FEMALE │\n", - "│ Gentoo ┆ Biscoe ┆ 49.9 ┆ 16.1 ┆ 213 ┆ 5400 ┆ MALE │\n", - "└─────────┴────────┴────────────────┴───────────────┴───────────────────┴─────────────┴────────┘" - ] - }, - "execution_count": 43, - "metadata": {}, - "output_type": "execute_result" - } + "cells": [ + { + "cell_type": "markdown", + "id": "6f248326", + "metadata": {}, + "source": [ + "# Getting Started with Python: Exploratory Data Analysis (Polars)\n", + "\n", + "**Author:** Alp Tezbasaran \n", + "**Updated by:** Bhaskar Ray\n", + "**Last updated:** 2026-06-26 \n", + "\n", + "This notebook is designed for a **live, in-person workshop**. It is intentionally iterative: we’ll ask a question, inspect the data, try a quick transformation or plot, then repeat.\n" + ] + }, + { + "cell_type": "markdown", + "id": "c5cb57c1", + "metadata": {}, + "source": [ + "## 0. Accessibility: theme, text size, and output\n", + "\n", + "- **Tools → Settings → Theme**: Light / Dark / System\n", + "- **Tools → Settings → Editor → Font size**: increase if needed\n", + "- **Browser zoom**: Cmd/Ctrl `+` or `-`\n", + "\n", + "Teaching tip: agree on a standard zoom level at the start.\n" + ] + }, + { + "cell_type": "markdown", + "id": "b3dafe05", + "metadata": {}, + "source": [ + "## Learning outcomes\n", + "\n", + "By the end of this workshop, you will be able to:\n", + "\n", + "- Load tabular data into Polars\n", + "- Profile data (shape, schema, missingness, unique values)\n", + "- Use **expressions** to filter and create features\n", + "- Summarize with `group_by().agg()` and explain what information is lost\n", + "- Use pivoting as **“groupby + aggregation + reshaping”**\n", + "- Make quick univariate and bivariate plots for exploration\n", + "- Recognize when to use **lazy** execution for performance\n" + ] + }, + { + "cell_type": "markdown", + "id": "25e87d09", + "metadata": {}, + "source": [ + "## The EDA loop (a simple roadmap)\n", + "\n", + "EDA is not linear. A useful loop is:\n", + "\n", + "1. **Inventory**: What columns do we have? What types? How big?\n", + "2. **Quality**: Missing values, duplicates, weird values\n", + "3. **Univariate**: What does each variable look like by itself?\n", + "4. **Bivariate / multivariate**: What relationships appear?\n", + "5. **Feature ideas**: Create features that clarify relationships\n", + "6. **Summarize & communicate**: tables/plots + short written takeaways\n", + "\n", + "```mermaid\n", + "flowchart LR\n", + " A[Inventory] --> B[Quality]\n", + " B --> C[Univariate]\n", + " C --> D[Bivariate / multivariate]\n", + " D --> E[Feature ideas]\n", + " E --> F[Summarize & communicate]\n", + " F --> A\n", + "```\n", + "\n", + "We’ll follow that loop on a few small datasets.\n" + ] + }, + { + "cell_type": "markdown", + "id": "b84b21f5", + "metadata": {}, + "source": [ + "## 1. Setup\n", + "\n", + "We’ll use **Polars** for data work and **matplotlib** for plotting.\n", + "\n", + "If you’re in Google Colab, install Polars once per session.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "9e1c1683", + "metadata": {}, + "outputs": [], + "source": [ + "import polars as pl\n", + "import matplotlib.pyplot as plt\n", + "from pathlib import Path\n", + "import numpy as np\n", + "\n", + "data_dir = Path('data')\n", + "\n", + "plt.rcParams['figure.dpi'] = 120\n", + "plt.rcParams['axes.grid'] = True" + ] + }, + { + "cell_type": "markdown", + "id": "f33be526", + "metadata": {}, + "source": [ + "## Polars: the 3 practical pillars\n", + "\n", + "Think of Polars as: **operation + expression + execution mode**.\n", + "\n", + "This gives students a simple mental model:\n", + "\n", + "1. **Operation (what you do)**\n", + " - DataFrame/LazyFrame methods define the step: `select`, `filter`, `with_columns`, `group_by`, `join`.\n", + "2. **Expression (how you describe it)**\n", + " - Expressions define the rule/math: `pl.col(...)`, `pl.when(...).then(...).otherwise(...)`, boolean logic, arithmetic.\n", + "3. **Execution mode (when/how it runs)**\n", + " - Eager runs now.\n", + " - Lazy builds a plan and runs at `collect()`, which helps Polars optimize.\n", + "\n", + "```mermaid\n", + "flowchart LR\n", + " A[Operation: select/filter/group_by] --> B[Expression: pl.col + rules]\n", + " B --> C[Execution mode: eager or lazy]\n", + "```\n", + "\n", + "### Easy analogy: cooking\n", + "\n", + "- **Operation** = the cooking action (chop, mix, bake)\n", + "- **Expression** = the recipe rule (\"add 2 cups\", \"if too salty, add water\")\n", + "- **Execution mode** = when you cook\n", + " - eager: cook each step immediately\n", + " - lazy: plan the full recipe first, then cook efficiently\n", + "\n", + "For today, we focus on the first two pillars most of the time.\n", + "We introduce lazy later as a performance bonus, not a required concept.\n" + ] + }, + { + "cell_type": "markdown", + "id": "1d2a27ac", + "metadata": {}, + "source": [ + "### Expression primer (5 minutes)\n", + "\n", + "In Polars, you rarely write row-by-row loops. You write **column expressions**.\n", + "\n", + "Common building blocks:\n", + "- `pl.col('x')` → refer to a column\n", + "- `.alias('new_name')` → rename an expression result\n", + "- Boolean logic: `&` (and), `|` (or), `~` (not)\n", + "- **Use parentheses** with boolean logic: `(cond1) & (cond2)`\n", + "- Conditional: `pl.when(cond).then(val).otherwise(val)`\n", + "- String/date helpers: `.str.*`, `.dt.*`\n" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "2f603b66", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (4, 5)
xsx_filleds_cleanflag
i64stri64strstr
1" A "1"a""ok"
2"b"2"b""ok"
3"b"3"b""ok"
nullnull0"missing""was_null"
" ], - "source": [ - "penguins_raw.tail(5)" - ] - }, - { - "cell_type": "markdown", - "id": "b03ef085", - "metadata": {}, - "source": [ - "### Missingness: counts and rates\n", - "\n", - "`null_count()` gives counts. For teaching and comparison across tables, rates (percent) are often easier to interpret.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 44, - "id": "5a4f602f", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (12, 4)
columnnull_countnull_ratenull_pct
stru32f64f64
"Post Employment Position"470.9494.0
"Termination Year"460.9292.0
"Termination Date"460.9292.0
"Termination Reason"430.8686.0
"Hire Year"380.7676.0
…………
"Nick Name"290.5858.0
"Fur Color"170.3434.0
"Prior Position"90.1818.0
"Prior Employer"90.1818.0
"Breed"10.022.0
" - ], - "text/plain": [ - "shape: (12, 4)\n", - "┌──────────────────────────┬────────────┬───────────┬──────────┐\n", - "│ column ┆ null_count ┆ null_rate ┆ null_pct │\n", - "│ --- ┆ --- ┆ --- ┆ --- │\n", - "│ str ┆ u32 ┆ f64 ┆ f64 │\n", - "╞══════════════════════════╪════════════╪═══════════╪══════════╡\n", - "│ Post Employment Position ┆ 47 ┆ 0.94 ┆ 94.0 │\n", - "│ Termination Year ┆ 46 ┆ 0.92 ┆ 92.0 │\n", - "│ Termination Date ┆ 46 ┆ 0.92 ┆ 92.0 │\n", - "│ Termination Reason ┆ 43 ┆ 0.86 ┆ 86.0 │\n", - "│ Hire Year ┆ 38 ┆ 0.76 ┆ 76.0 │\n", - "│ … ┆ … ┆ … ┆ … │\n", - "│ Nick Name ┆ 29 ┆ 0.58 ┆ 58.0 │\n", - "│ Fur Color ┆ 17 ┆ 0.34 ┆ 34.0 │\n", - "│ Prior Position ┆ 9 ┆ 0.18 ┆ 18.0 │\n", - "│ Prior Employer ┆ 9 ┆ 0.18 ┆ 18.0 │\n", - "│ Breed ┆ 1 ┆ 0.02 ┆ 2.0 │\n", - "└──────────────────────────┴────────────┴───────────┴──────────┘" - ] - }, - "execution_count": 44, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (4, 5)\n", + "┌──────┬──────┬──────────┬─────────┬──────────┐\n", + "│ x ┆ s ┆ x_filled ┆ s_clean ┆ flag │\n", + "│ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", + "│ i64 ┆ str ┆ i64 ┆ str ┆ str │\n", + "╞══════╪══════╪══════════╪═════════╪══════════╡\n", + "│ 1 ┆ A ┆ 1 ┆ a ┆ ok │\n", + "│ 2 ┆ b ┆ 2 ┆ b ┆ ok │\n", + "│ 3 ┆ b ┆ 3 ┆ b ┆ ok │\n", + "│ null ┆ null ┆ 0 ┆ missing ┆ was_null │\n", + "└──────┴──────┴──────────┴─────────┴──────────┘" + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "expr_demo = pl.DataFrame({'x':[1,2,3,None], 's':[' A ','b','b',None]})\n", + "(\n", + " expr_demo\n", + " .with_columns([\n", + " pl.col('x').fill_null(0).alias('x_filled'),\n", + " pl.col('s').str.to_lowercase().str.strip_chars().fill_null('missing').alias('s_clean'),\n", + " pl.when(pl.col('x').is_null()).then(pl.lit('was_null')).otherwise(pl.lit('ok')).alias('flag')\n", + " ])\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "969f55dd", + "metadata": {}, + "source": [ + "## Checkpoint A (quick)\n", + "\n", + "- Everyone can run the setup cell and the expression demo\n", + "- If you’re stuck on imports/paths, follow along and we’ll fix it during the next pause\n" + ] + }, + { + "cell_type": "markdown", + "id": "f2839e6f", + "metadata": {}, + "source": [ + "## 2. Load datasets\n", + "\n", + "We’ll use three small datasets:\n", + "\n", + "- `NCSU_Mascots_v1.csv` *(synthetic)*\n", + "- `NCSU Celebrity Graduates_v1.csv` *(synthetic)*\n", + "- `penguins.csv` *(public dataset, light cleaning demo)*\n", + "\n", + "Even with synthetic data, practice good habits: avoid re-identification and think about bias.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "0e4e3911", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "((50, 22), (100, 16), (344, 7))" + ] + }, + "execution_count": 3, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "mascots = pl.read_csv(data_dir / 'NCSU_Mascots_v1.csv')\n", + "celebs = pl.read_csv(data_dir / 'NCSU Celebrity Graduates_v1.csv')\n", + "penguins_raw = pl.read_csv(data_dir / 'penguins.csv')\n", + "\n", + "(mascots.shape, celebs.shape, penguins_raw.shape)" + ] + }, + { + "cell_type": "markdown", + "id": "3b93421f", + "metadata": {}, + "source": [ + "## 3. Inventory & profiling\n", + "\n", + "Start every EDA with:\n", + "- **shape** (rows, columns)\n", + "- **schema** (column types)\n", + "- **preview** (head/tail)\n", + "\n", + "Then a quick profile:\n", + "- missingness (counts *and* rates)\n", + "- numeric summaries\n", + "- categorical cardinality (how many unique values?)\n", + "- duplicates (do we expect any?)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "773bf757", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Mascots: (50, 22)\n", + "Schema({'Name': String, 'Species': String, 'Breed': String, 'Prior Position': String, 'Prior Employer': String, 'Interview Date': String, 'Hire Year': Int64, 'Hire Date': String, 'Termination Year': Float64, 'Termination Date': String, 'Termination Reason': String, 'Post Employment Position': String, 'Nick Name': String, 'Temperment': String, 'Fur Color': String, 'Age': Float64, 'Height': Float64, 'Weight': String, 'Offer Sent': Int64, 'Starting Salary': Int64, 'Biter': Int64, 'Hired': Int64})\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "shape: (5, 22)
NameSpeciesBreedPrior PositionPrior EmployerInterview DateHire YearHire DateTermination YearTermination DateTermination ReasonPost Employment PositionNick NameTempermentFur ColorAgeHeightWeightOffer SentStarting SalaryBiterHired
strstrstrstrstrstri64strf64strstrstrstrstrstrf64f64stri64i64i64i64
"Wallace Whitfield Riddick""Human"nullnullnull"1899-02-17"1899"1899-02-23"nullnullnullnullnull"patient""cream"5.034.5"40.0"1948101
"Tige""Dog""English Bulldog"nullnull"1910-02-16"1910"1910-02-17"nullnullnullnullnull"sweet""brown"10.914.2"51.0"11013001
"Togo""Dog""American Bulldog"nullnull"1910-06-28"1910"1910-07-01"nullnullnullnullnull"sweet""brown"10.424.4"90.0"1957201
"State""Wolf""Timber Wolf"nullnull"1940-05-02"1940"1940-05-03"1946.0"1940-05-17""New Job""Traveling Zoo"null"bewildered""multi-color"9.829.8"112.5"1992101
"Lobo I""Wolf""Timber Wolf""Zoo""Philadelphia Zoo""1959-08-26"1959"1959-09-03"1959.0"1959-05-29""Passed Away"nullnull"quiet""gray"10.327.2"100.15"1950901
" ], - "source": [ - "n = mascots.height\n", - "\n", - "mascots_missingness = (\n", - " mascots\n", - " .null_count()\n", - " .transpose(include_header=True, header_name='column', column_names=['null_count'])\n", - " .with_columns([\n", - " (pl.col('null_count') / pl.lit(n)).alias('null_rate'),\n", - " (pl.col('null_count') / pl.lit(n) * 100).alias('null_pct'),\n", - " ])\n", - " .sort('null_count', descending=True)\n", - ")\n", - "\n", - "mascots_missingness.head(12)" - ] - }, - { - "cell_type": "markdown", - "id": "b4a973a0", - "metadata": {}, - "source": [ - "### Quick numeric summary\n", - "\n", - "This is triage, not a final report.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 45, - "id": "5bf931c8", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (9, 23)
statisticNameSpeciesBreedPrior PositionPrior EmployerInterview DateHire YearHire DateTermination YearTermination DateTermination ReasonPost Employment PositionNick NameTempermentFur ColorAgeHeightWeightOffer SentStarting SalaryBiterHired
strstrstrstrstrstrstrf64strf64strstrstrstrstrstrf64f64strf64f64f64f64
"count""50""50""49""41""41""50"12.0"12"4.0"4""7""3""21""49""33"50.050.0"50"50.017.050.050.0
"null_count""0""0""1""9""9""0"38.0"38"46.0"46""43""47""29""1""17"0.00.0"0"0.033.00.00.0
"mean"nullnullnullnullnullnull1964.25null1973.75nullnullnullnullnullnull10.32419.284null0.349862.7647060.080.24
"std"nullnullnullnullnullnull43.587164null32.356092nullnullnullnullnullnull1.91626316.461679null0.478518356.5078070.2740480.431419
"min""Ash""Alpaca""African Grey Parrot"" Demolition Expert""ACME Food Storage""1899-02-17"1899.0"1899-02-23"1946.0"1940-05-17""Escaped""Art Thief""Boo Boo""bewildered""black"5.01.25"0.0003"0.09313.00.00.0
"25%"nullnullnullnullnullnull1940.0null1959.0nullnullnullnullnullnull9.17.59null0.09572.00.00.0
"50%"nullnullnullnullnullnull1966.0null1970.0nullnullnullnullnullnull10.313.25null0.09904.00.00.0
"75%"nullnullnullnullnullnull2010.0null1970.0nullnullnullnullnullnull11.626.2null1.010086.00.00.0
"max""Willow""Wolf""Wolf/German Shepard""Zoo""Walmart""2024-09-09"2021.0"2021-11-17"2020.0"2020-11-09""Retired""Traveling Zoo""Yoda""sweet""white"14.777.34"90.0"1.010478.01.01.0
" - ], - "text/plain": [ - "shape: (9, 23)\n", - "┌────────────┬────────┬─────────┬─────────────┬───┬────────────┬─────────────┬──────────┬──────────┐\n", - "│ statistic ┆ Name ┆ Species ┆ Breed ┆ … ┆ Offer Sent ┆ Starting ┆ Biter ┆ Hired │\n", - "│ --- ┆ --- ┆ --- ┆ --- ┆ ┆ --- ┆ Salary ┆ --- ┆ --- │\n", - "│ str ┆ str ┆ str ┆ str ┆ ┆ f64 ┆ --- ┆ f64 ┆ f64 │\n", - "│ ┆ ┆ ┆ ┆ ┆ ┆ f64 ┆ ┆ │\n", - "╞════════════╪════════╪═════════╪═════════════╪═══╪════════════╪═════════════╪══════════╪══════════╡\n", - "│ count ┆ 50 ┆ 50 ┆ 49 ┆ … ┆ 50.0 ┆ 17.0 ┆ 50.0 ┆ 50.0 │\n", - "│ null_count ┆ 0 ┆ 0 ┆ 1 ┆ … ┆ 0.0 ┆ 33.0 ┆ 0.0 ┆ 0.0 │\n", - "│ mean ┆ null ┆ null ┆ null ┆ … ┆ 0.34 ┆ 9862.764706 ┆ 0.08 ┆ 0.24 │\n", - "│ std ┆ null ┆ null ┆ null ┆ … ┆ 0.478518 ┆ 356.507807 ┆ 0.274048 ┆ 0.431419 │\n", - "│ min ┆ Ash ┆ Alpaca ┆ African ┆ … ┆ 0.0 ┆ 9313.0 ┆ 0.0 ┆ 0.0 │\n", - "│ ┆ ┆ ┆ Grey Parrot ┆ ┆ ┆ ┆ ┆ │\n", - "│ 25% ┆ null ┆ null ┆ null ┆ … ┆ 0.0 ┆ 9572.0 ┆ 0.0 ┆ 0.0 │\n", - "│ 50% ┆ null ┆ null ┆ null ┆ … ┆ 0.0 ┆ 9904.0 ┆ 0.0 ┆ 0.0 │\n", - "│ 75% ┆ null ┆ null ┆ null ┆ … ┆ 1.0 ┆ 10086.0 ┆ 0.0 ┆ 0.0 │\n", - "│ max ┆ Willow ┆ Wolf ┆ Wolf/German ┆ … ┆ 1.0 ┆ 10478.0 ┆ 1.0 ┆ 1.0 │\n", - "│ ┆ ┆ ┆ Shepard ┆ ┆ ┆ ┆ ┆ │\n", - "└────────────┴────────┴─────────┴─────────────┴───┴────────────┴─────────────┴──────────┴──────────┘" - ] - }, - "execution_count": 45, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (5, 22)\n", + "┌───────────┬─────────┬──────────────────┬──────────┬───┬────────────┬──────────┬───────┬───────┐\n", + "│ Name ┆ Species ┆ Breed ┆ Prior ┆ … ┆ Offer Sent ┆ Starting ┆ Biter ┆ Hired │\n", + "│ --- ┆ --- ┆ --- ┆ Position ┆ ┆ --- ┆ Salary ┆ --- ┆ --- │\n", + "│ str ┆ str ┆ str ┆ --- ┆ ┆ i64 ┆ --- ┆ i64 ┆ i64 │\n", + "│ ┆ ┆ ┆ str ┆ ┆ ┆ i64 ┆ ┆ │\n", + "╞═══════════╪═════════╪══════════════════╪══════════╪═══╪════════════╪══════════╪═══════╪═══════╡\n", + "│ Wallace ┆ Human ┆ null ┆ null ┆ … ┆ 1 ┆ 9481 ┆ 0 ┆ 1 │\n", + "│ Whitfield ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │\n", + "│ Riddick ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │\n", + "│ Tige ┆ Dog ┆ English Bulldog ┆ null ┆ … ┆ 1 ┆ 10130 ┆ 0 ┆ 1 │\n", + "│ Togo ┆ Dog ┆ American Bulldog ┆ null ┆ … ┆ 1 ┆ 9572 ┆ 0 ┆ 1 │\n", + "│ State ┆ Wolf ┆ Timber Wolf ┆ null ┆ … ┆ 1 ┆ 9921 ┆ 0 ┆ 1 │\n", + "│ Lobo I ┆ Wolf ┆ Timber Wolf ┆ Zoo ┆ … ┆ 1 ┆ 9509 ┆ 0 ┆ 1 │\n", + "└───────────┴─────────┴──────────────────┴──────────┴───┴────────────┴──────────┴───────┴───────┘" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "print('Mascots:', mascots.shape)\n", + "print(mascots.schema)\n", + "mascots.head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "279d69ce", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (5, 22)
NameSpeciesBreedPrior PositionPrior EmployerInterview DateHire YearHire DateTermination YearTermination DateTermination ReasonPost Employment PositionNick NameTempermentFur ColorAgeHeightWeightOffer SentStarting SalaryBiterHired
strstrstrstrstrstri64strf64strstrstrstrstrstrf64f64stri64i64i64i64
"Shadow""Spider""Golden Orb Weaver""Web Infrastructure Designer""Giggle ""2013-04-12"nullnullnullnullnullnull"Gizmo""devilishly clever"null11.44.0"0.009"0null00
"Clover""Rabbit""Mini Rex""Burrow Engineer""Hillside Civil Engineering""1952-07-30"nullnullnullnullnullnullnull"frightened""brown"10.012.4"3.71"0null00
"Sophie""Rabbit""Holland Lop""Carrot Quality Control Associa…"Publix""1953-11-17"nullnullnullnullnullnull"Spock""social""black"9.710.55"3.18"1990400
"Pixel""Turtle""Eastern Box""Rock Impersonator""No Moves, LLC""1997-08-25"nullnullnullnullnullnull"Gizmo""patient"null11.65.44"1.49"0null00
"Baron""Dog""Dalmatian""Firehouse Morale Officer""Raleigh Fire Department""1990-08-06"nullnullnullnullnullnullnull"social""black"14.721.5"57.68"0null00
" ], - "source": [ - "mascots.describe()" - ] - }, - { - "cell_type": "markdown", - "id": "26e4a94f", - "metadata": {}, - "source": [ - "### Categorical “shape” (unique counts)\n", - "\n", - "This is a fast way to find columns with many categories (IDs, free text, etc.).\n" - ] - }, - { - "cell_type": "code", - "execution_count": 46, - "id": "e6ccf641", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (22, 2)
columnn_unique
stru32
"Name"50
"Interview Date"50
"Weight"50
"Height"49
"Breed"45
……
"Termination Date"5
"Post Employment Position"4
"Offer Sent"2
"Biter"2
"Hired"2
" - ], - "text/plain": [ - "shape: (22, 2)\n", - "┌──────────────────────────┬──────────┐\n", - "│ column ┆ n_unique │\n", - "│ --- ┆ --- │\n", - "│ str ┆ u32 │\n", - "╞══════════════════════════╪══════════╡\n", - "│ Name ┆ 50 │\n", - "│ Interview Date ┆ 50 │\n", - "│ Weight ┆ 50 │\n", - "│ Height ┆ 49 │\n", - "│ Breed ┆ 45 │\n", - "│ … ┆ … │\n", - "│ Termination Date ┆ 5 │\n", - "│ Post Employment Position ┆ 4 │\n", - "│ Offer Sent ┆ 2 │\n", - "│ Biter ┆ 2 │\n", - "│ Hired ┆ 2 │\n", - "└──────────────────────────┴──────────┘" - ] - }, - "execution_count": 46, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (5, 22)\n", + "┌────────┬─────────┬─────────────┬─────────────────────┬───┬────────────┬──────────┬───────┬───────┐\n", + "│ Name ┆ Species ┆ Breed ┆ Prior Position ┆ … ┆ Offer Sent ┆ Starting ┆ Biter ┆ Hired │\n", + "│ --- ┆ --- ┆ --- ┆ --- ┆ ┆ --- ┆ Salary ┆ --- ┆ --- │\n", + "│ str ┆ str ┆ str ┆ str ┆ ┆ i64 ┆ --- ┆ i64 ┆ i64 │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ i64 ┆ ┆ │\n", + "╞════════╪═════════╪═════════════╪═════════════════════╪═══╪════════════╪══════════╪═══════╪═══════╡\n", + "│ Shadow ┆ Spider ┆ Golden Orb ┆ Web Infrastructure ┆ … ┆ 0 ┆ null ┆ 0 ┆ 0 │\n", + "│ ┆ ┆ Weaver ┆ Designer ┆ ┆ ┆ ┆ ┆ │\n", + "│ Clover ┆ Rabbit ┆ Mini Rex ┆ Burrow Engineer ┆ … ┆ 0 ┆ null ┆ 0 ┆ 0 │\n", + "│ Sophie ┆ Rabbit ┆ Holland Lop ┆ Carrot Quality ┆ … ┆ 1 ┆ 9904 ┆ 0 ┆ 0 │\n", + "│ ┆ ┆ ┆ Control Associa… ┆ ┆ ┆ ┆ ┆ │\n", + "│ Pixel ┆ Turtle ┆ Eastern Box ┆ Rock Impersonator ┆ … ┆ 0 ┆ null ┆ 0 ┆ 0 │\n", + "│ Baron ┆ Dog ┆ Dalmatian ┆ Firehouse Morale ┆ … ┆ 0 ┆ null ┆ 0 ┆ 0 │\n", + "│ ┆ ┆ ┆ Officer ┆ ┆ ┆ ┆ ┆ │\n", + "└────────┴─────────┴─────────────┴─────────────────────┴───┴────────────┴──────────┴───────┴───────┘" + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "mascots.tail(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "708a7be5", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Celebs: (100, 16)\n", + "Schema({'#': Int64, 'Name': String, 'Birth Year': Int64, 'Death Year': Float64, 'Occupation(s)': String, 'Country Born In': String, 'Birthdate': String, 'Age': Int64, 'Graduation Month/Year': String, 'Workstudy Position': String, 'Workstudy Hourly Rate': Float64, 'GPA': Float64, 'College': String, 'Loved Library?': Int64, 'Favorite NCSU Hangout': String, 'Favorite Raleigh Hangout': String})\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "shape: (5, 16)
#NameBirth YearDeath YearOccupation(s)Country Born InBirthdateAgeGraduation Month/YearWorkstudy PositionWorkstudy Hourly RateGPACollegeLoved Library?Favorite NCSU HangoutFavorite Raleigh Hangout
i64stri64f64strstrstri64strstrf64f64stri64strstr
88"Adrian Wilson"1979null"American football player""United States""1979-10-12"45"Dec/1997""IT"15.523.88"Humanities and Social Sciences"0"Library""Cup A Joe"
95"Ahmet Özal"1955null"businessperson, politician""Turkey""1955-09-05"69"Dec/1974""Campus Events"16.443.54"Education"0"Fountain Dining Hall""The Optimist"
72"Akihiro Kitamura"1979null"seiyū, actor, television actor""Japan""1979-03-25"45"Dec/1997""IT"16.892.17"Natural Resources"0"Free Expression Tunnel""The Optimist"
10"Anna Camp"1982null"film actor, television actor, …"United States""1982-09-27"42"Dec/2000""Campus Events"14.842.18"Veterinary Medicine"0"Fountain Dining Hall""Mitch's Tavern"
5"Anthony Mackie"1978null"stage actor, film actor, telev…"United States""1978-09-23"46"Dec/1996""Library"17.793.55"Textiles"0"The Brickyard""State Farmers Market Restauran…
" ], - "source": [ - "(\n", - " mascots\n", - " .select(pl.all().n_unique())\n", - " .transpose(include_header=True, header_name='column', column_names=['n_unique'])\n", - " .sort('n_unique', descending=True)\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "11ab2d30", - "metadata": {}, - "source": [ - "### Duplicates (quick check)\n", - "\n", - "First ask: **do we expect duplicates?** If not, check for them.\n", - "\n", - "Note: “duplicate” depends on *which columns* define identity.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 47, - "id": "06ecb5d7", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (1, 2)
n_rowsn_duplicate_rows
u32u32
500
" - ], - "text/plain": [ - "shape: (1, 2)\n", - "┌────────┬──────────────────┐\n", - "│ n_rows ┆ n_duplicate_rows │\n", - "│ --- ┆ --- │\n", - "│ u32 ┆ u32 │\n", - "╞════════╪══════════════════╡\n", - "│ 50 ┆ 0 │\n", - "└────────┴──────────────────┘" - ] - }, - "execution_count": 47, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (5, 16)\n", + "┌─────┬─────────────┬───────┬─────────────┬───┬─────────────┬────────────┬────────────┬────────────┐\n", + "│ # ┆ Name ┆ Birth ┆ Death Year ┆ … ┆ College ┆ Loved ┆ Favorite ┆ Favorite │\n", + "│ --- ┆ --- ┆ Year ┆ --- ┆ ┆ --- ┆ Library? ┆ NCSU ┆ Raleigh │\n", + "│ i64 ┆ str ┆ --- ┆ f64 ┆ ┆ str ┆ --- ┆ Hangout ┆ Hangout │\n", + "│ ┆ ┆ i64 ┆ ┆ ┆ ┆ i64 ┆ --- ┆ --- │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ str ┆ str │\n", + "╞═════╪═════════════╪═══════╪═════════════╪═══╪═════════════╪════════════╪════════════╪════════════╡\n", + "│ 88 ┆ Adrian ┆ 1979 ┆ null ┆ … ┆ Humanities ┆ 0 ┆ Library ┆ Cup A Joe │\n", + "│ ┆ Wilson ┆ ┆ ┆ ┆ and Social ┆ ┆ ┆ │\n", + "│ ┆ ┆ ┆ ┆ ┆ Sciences ┆ ┆ ┆ │\n", + "│ 95 ┆ Ahmet Özal ┆ 1955 ┆ null ┆ … ┆ Education ┆ 0 ┆ Fountain ┆ The │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Dining ┆ Optimist │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Hall ┆ │\n", + "│ 72 ┆ Akihiro ┆ 1979 ┆ null ┆ … ┆ Natural ┆ 0 ┆ Free ┆ The │\n", + "│ ┆ Kitamura ┆ ┆ ┆ ┆ Resources ┆ ┆ Expression ┆ Optimist │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Tunnel ┆ │\n", + "│ 10 ┆ Anna Camp ┆ 1982 ┆ null ┆ … ┆ Veterinary ┆ 0 ┆ Fountain ┆ Mitch's │\n", + "│ ┆ ┆ ┆ ┆ ┆ Medicine ┆ ┆ Dining ┆ Tavern │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Hall ┆ │\n", + "│ 5 ┆ Anthony ┆ 1978 ┆ null ┆ … ┆ Textiles ┆ 0 ┆ The ┆ State │\n", + "│ ┆ Mackie ┆ ┆ ┆ ┆ ┆ ┆ Brickyard ┆ Farmers │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Market │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Restauran… │\n", + "└─────┴─────────────┴───────┴─────────────┴───┴─────────────┴────────────┴────────────┴────────────┘" + ] + }, + "execution_count": 6, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "print('Celebs:', celebs.shape)\n", + "print(celebs.schema)\n", + "celebs.head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "0b3cc41e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (5, 16)
#NameBirth YearDeath YearOccupation(s)Country Born InBirthdateAgeGraduation Month/YearWorkstudy PositionWorkstudy Hourly RateGPACollegeLoved Library?Favorite NCSU HangoutFavorite Raleigh Hangout
i64stri64f64strstrstri64strstrf64f64stri64strstr
70"Vinny Del Negro"1966null"basketball player, basketball …"United States""1966-08-09"58"Dec/1984""Dining"15.93.49"Sciences"1"The Brickyard""Jubala Coffee"
69"Vivian Howard"1978null"restaurateur, cook"null"1978-03-18"46"Dec/1996""Dining"17.223.37"Humanities and Social Sciences"0"The Brickyard""Raleigh Beer Garden"
90"Walter B. Jones"19432019.0"politician""United States""1943-02-10"81"Dec/1962""Campus Events"17.433.46"Design"0"Reynolds Coliseum""Irish Coffee Lab"
65"Yasonna Laoly"1953null"politician, political candidat…"Indonesia""1953-05-27"71"Dec/1972""Theater"16.892.57"Natural Resources"0"Free Expression Tunnel""Boxcar"
6"Zach Galifianakis"1969null"voice actor, screen writer, fi…"United States""1969-10-01"55"May/1987""Dining"14.242.42"Design"1"The Brickyard""The Optimist"
" ], - "source": [ - "mascots.select([\n", - " pl.len().alias('n_rows'),\n", - " pl.struct(pl.all()).is_duplicated().sum().alias('n_duplicate_rows')\n", - "])" - ] - }, - { - "cell_type": "markdown", - "id": "0a60b4cf", - "metadata": {}, - "source": [ - "### Try it yourself (profiling)\n", - "\n", - "- **Minimum**: show the 10 columns with the highest `null_count` in `mascots`.\n", - "- **Stretch**: show the 10 columns with the highest `null_pct`.\n", - "\n", - "Expected result shape: a table with columns like `column`, `null_count`, `null_rate`, `null_pct`.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 48, - "id": "b0c6cf5d", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (10, 4)
columnnull_countnull_ratenull_pct
stru32f64f64
"Post Employment Position"470.9494.0
"Termination Year"460.9292.0
"Termination Date"460.9292.0
"Termination Reason"430.8686.0
"Hire Year"380.7676.0
"Hire Date"380.7676.0
"Starting Salary"330.6666.0
"Nick Name"290.5858.0
"Fur Color"170.3434.0
"Prior Position"90.1818.0
" - ], - "text/plain": [ - "shape: (10, 4)\n", - "┌──────────────────────────┬────────────┬───────────┬──────────┐\n", - "│ column ┆ null_count ┆ null_rate ┆ null_pct │\n", - "│ --- ┆ --- ┆ --- ┆ --- │\n", - "│ str ┆ u32 ┆ f64 ┆ f64 │\n", - "╞══════════════════════════╪════════════╪═══════════╪══════════╡\n", - "│ Post Employment Position ┆ 47 ┆ 0.94 ┆ 94.0 │\n", - "│ Termination Year ┆ 46 ┆ 0.92 ┆ 92.0 │\n", - "│ Termination Date ┆ 46 ┆ 0.92 ┆ 92.0 │\n", - "│ Termination Reason ┆ 43 ┆ 0.86 ┆ 86.0 │\n", - "│ Hire Year ┆ 38 ┆ 0.76 ┆ 76.0 │\n", - "│ Hire Date ┆ 38 ┆ 0.76 ┆ 76.0 │\n", - "│ Starting Salary ┆ 33 ┆ 0.66 ┆ 66.0 │\n", - "│ Nick Name ┆ 29 ┆ 0.58 ┆ 58.0 │\n", - "│ Fur Color ┆ 17 ┆ 0.34 ┆ 34.0 │\n", - "│ Prior Position ┆ 9 ┆ 0.18 ┆ 18.0 │\n", - "└──────────────────────────┴────────────┴───────────┴──────────┘" - ] - }, - "execution_count": 48, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (5, 16)\n", + "┌─────┬─────────────┬───────┬─────────────┬───┬─────────────┬────────────┬────────────┬────────────┐\n", + "│ # ┆ Name ┆ Birth ┆ Death Year ┆ … ┆ College ┆ Loved ┆ Favorite ┆ Favorite │\n", + "│ --- ┆ --- ┆ Year ┆ --- ┆ ┆ --- ┆ Library? ┆ NCSU ┆ Raleigh │\n", + "│ i64 ┆ str ┆ --- ┆ f64 ┆ ┆ str ┆ --- ┆ Hangout ┆ Hangout │\n", + "│ ┆ ┆ i64 ┆ ┆ ┆ ┆ i64 ┆ --- ┆ --- │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ str ┆ str │\n", + "╞═════╪═════════════╪═══════╪═════════════╪═══╪═════════════╪════════════╪════════════╪════════════╡\n", + "│ 70 ┆ Vinny Del ┆ 1966 ┆ null ┆ … ┆ Sciences ┆ 1 ┆ The ┆ Jubala │\n", + "│ ┆ Negro ┆ ┆ ┆ ┆ ┆ ┆ Brickyard ┆ Coffee │\n", + "│ 69 ┆ Vivian ┆ 1978 ┆ null ┆ … ┆ Humanities ┆ 0 ┆ The ┆ Raleigh │\n", + "│ ┆ Howard ┆ ┆ ┆ ┆ and Social ┆ ┆ Brickyard ┆ Beer │\n", + "│ ┆ ┆ ┆ ┆ ┆ Sciences ┆ ┆ ┆ Garden │\n", + "│ 90 ┆ Walter B. ┆ 1943 ┆ 2019.0 ┆ … ┆ Design ┆ 0 ┆ Reynolds ┆ Irish │\n", + "│ ┆ Jones ┆ ┆ ┆ ┆ ┆ ┆ Coliseum ┆ Coffee Lab │\n", + "│ 65 ┆ Yasonna ┆ 1953 ┆ null ┆ … ┆ Natural ┆ 0 ┆ Free ┆ Boxcar │\n", + "│ ┆ Laoly ┆ ┆ ┆ ┆ Resources ┆ ┆ Expression ┆ │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ ┆ Tunnel ┆ │\n", + "│ 6 ┆ Zach Galifi ┆ 1969 ┆ null ┆ … ┆ Design ┆ 1 ┆ The ┆ The │\n", + "│ ┆ anakis ┆ ┆ ┆ ┆ ┆ ┆ Brickyard ┆ Optimist │\n", + "└─────┴─────────────┴───────┴─────────────┴───┴─────────────┴────────────┴────────────┴────────────┘" + ] + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "celebs.tail(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "ddb32ec6", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Penguins: (344, 7)\n", + "Schema({'species': String, 'island': String, 'bill_length_mm': Float64, 'bill_depth_mm': Float64, 'flipper_length_mm': Int64, 'body_mass_g': Int64, 'sex': String})\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "shape: (5, 7)
speciesislandbill_length_mmbill_depth_mmflipper_length_mmbody_mass_gsex
strstrf64f64i64i64str
"Adelie""Torgersen"39.118.71813750"MALE"
"Adelie""Torgersen"39.517.41863800"FEMALE"
"Adelie""Torgersen"40.318.01953250"FEMALE"
"Adelie""Torgersen"nullnullnullnullnull
"Adelie""Torgersen"36.719.31933450"FEMALE"
" ], - "source": [ - "(\n", - " mascots\n", - " .null_count()\n", - " .transpose(include_header=True, header_name='column', column_names=['null_count'])\n", - " .with_columns([\n", - " (pl.col('null_count') / pl.lit(mascots.height)).alias('null_rate'),\n", - " (pl.col('null_count') / pl.lit(mascots.height) * 100).alias('null_pct'),\n", - " ])\n", - " .sort('null_count', descending=True)\n", - " .head(10)\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "6012baf7", - "metadata": {}, - "source": [ - "## Checkpoint B\n", - "\n", - "You should now have a reusable profiling checklist: shape, schema, preview, missingness (counts/rates), `describe`, `n_unique`, and “do we expect duplicates?”.\n" - ] - }, - { - "cell_type": "markdown", - "id": "621f6939", - "metadata": {}, - "source": [ - "## 4. Univariate exploration\n", - "\n", - "Univariate exploration answers: **“What does this variable look like by itself?”**\n", - "\n", - "- Numeric: distribution (histogram), outliers (boxplot)\n", - "- Categorical: counts (bar chart)\n", - "\n", - "We’ll do a quick categorical count on `mascots` and a histogram on `celebs` GPA.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 49, - "id": "c3f7e618", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (23, 2)
Specieslen
stru32
"Dog"10
"Cat"5
"Bird"4
"Lizard"3
"Horse"3
……
"Chicken"1
"Alpaca"1
"Mix"1
"Hedgehog"1
"Snail"1
" - ], - "text/plain": [ - "shape: (23, 2)\n", - "┌──────────┬─────┐\n", - "│ Species ┆ len │\n", - "│ --- ┆ --- │\n", - "│ str ┆ u32 │\n", - "╞══════════╪═════╡\n", - "│ Dog ┆ 10 │\n", - "│ Cat ┆ 5 │\n", - "│ Bird ┆ 4 │\n", - "│ Lizard ┆ 3 │\n", - "│ Horse ┆ 3 │\n", - "│ … ┆ … │\n", - "│ Chicken ┆ 1 │\n", - "│ Alpaca ┆ 1 │\n", - "│ Mix ┆ 1 │\n", - "│ Hedgehog ┆ 1 │\n", - "│ Snail ┆ 1 │\n", - "└──────────┴─────┘" - ] - }, - "execution_count": 49, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (5, 7)\n", + "┌─────────┬───────────┬────────────────┬───────────────┬───────────────────┬─────────────┬────────┐\n", + "│ species ┆ island ┆ bill_length_mm ┆ bill_depth_mm ┆ flipper_length_mm ┆ body_mass_g ┆ sex │\n", + "│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", + "│ str ┆ str ┆ f64 ┆ f64 ┆ i64 ┆ i64 ┆ str │\n", + "╞═════════╪═══════════╪════════════════╪═══════════════╪═══════════════════╪═════════════╪════════╡\n", + "│ Adelie ┆ Torgersen ┆ 39.1 ┆ 18.7 ┆ 181 ┆ 3750 ┆ MALE │\n", + "│ Adelie ┆ Torgersen ┆ 39.5 ┆ 17.4 ┆ 186 ┆ 3800 ┆ FEMALE │\n", + "│ Adelie ┆ Torgersen ┆ 40.3 ┆ 18.0 ┆ 195 ┆ 3250 ┆ FEMALE │\n", + "│ Adelie ┆ Torgersen ┆ null ┆ null ┆ null ┆ null ┆ null │\n", + "│ Adelie ┆ Torgersen ┆ 36.7 ┆ 19.3 ┆ 193 ┆ 3450 ┆ FEMALE │\n", + "└─────────┴───────────┴────────────────┴───────────────┴───────────────────┴─────────────┴────────┘" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "print('Penguins:', penguins_raw.shape)\n", + "print(penguins_raw.schema)\n", + "penguins_raw.head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "7ffb6cd5", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (5, 7)
speciesislandbill_length_mmbill_depth_mmflipper_length_mmbody_mass_gsex
strstrf64f64i64i64str
"Gentoo""Biscoe"nullnullnullnullnull
"Gentoo""Biscoe"46.814.32154850"FEMALE"
"Gentoo""Biscoe"50.415.72225750"MALE"
"Gentoo""Biscoe"45.214.82125200"FEMALE"
"Gentoo""Biscoe"49.916.12135400"MALE"
" ], - "source": [ - "species_counts = (\n", - " mascots\n", - " .group_by('Species')\n", - " .len()\n", - " .sort('len', descending=True)\n", - ")\n", - "species_counts" - ] - }, - { - "cell_type": "code", - "execution_count": 50, - "id": "43825124", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } + "text/plain": [ + "shape: (5, 7)\n", + "┌─────────┬────────┬────────────────┬───────────────┬───────────────────┬─────────────┬────────┐\n", + "│ species ┆ island ┆ bill_length_mm ┆ bill_depth_mm ┆ flipper_length_mm ┆ body_mass_g ┆ sex │\n", + "│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", + "│ str ┆ str ┆ f64 ┆ f64 ┆ i64 ┆ i64 ┆ str │\n", + "╞═════════╪════════╪════════════════╪═══════════════╪═══════════════════╪═════════════╪════════╡\n", + "│ Gentoo ┆ Biscoe ┆ null ┆ null ┆ null ┆ null ┆ null │\n", + "│ Gentoo ┆ Biscoe ┆ 46.8 ┆ 14.3 ┆ 215 ┆ 4850 ┆ FEMALE │\n", + "│ Gentoo ┆ Biscoe ┆ 50.4 ┆ 15.7 ┆ 222 ┆ 5750 ┆ MALE │\n", + "│ Gentoo ┆ Biscoe ┆ 45.2 ┆ 14.8 ┆ 212 ┆ 5200 ┆ FEMALE │\n", + "│ Gentoo ┆ Biscoe ┆ 49.9 ┆ 16.1 ┆ 213 ┆ 5400 ┆ MALE │\n", + "└─────────┴────────┴────────────────┴───────────────┴───────────────────┴─────────────┴────────┘" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "penguins_raw.tail(5)" + ] + }, + { + "cell_type": "markdown", + "id": "b03ef085", + "metadata": {}, + "source": [ + "### Missingness: counts and rates\n", + "\n", + "`null_count()` gives counts. For teaching and comparison across tables, rates (percent) are often easier to interpret.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "5a4f602f", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (12, 4)
columnnull_countnull_ratenull_pct
stru32f64f64
"Post Employment Position"470.9494.0
"Termination Year"460.9292.0
"Termination Date"460.9292.0
"Termination Reason"430.8686.0
"Hire Year"380.7676.0
…………
"Nick Name"290.5858.0
"Fur Color"170.3434.0
"Prior Position"90.1818.0
"Prior Employer"90.1818.0
"Breed"10.022.0
" ], - "source": [ - "top_n = 10\n", - "top_species = species_counts.head(top_n)\n", - "\n", - "plt.figure(figsize=(8, 4))\n", - "plt.bar(top_species['Species'].to_list(), top_species['len'].to_list())\n", - "plt.xticks(rotation=45, ha='right')\n", - "plt.ylabel('Count')\n", - "plt.title(f'Mascots by Species (top {top_n})')\n", - "plt.tight_layout()\n", - "plt.show()" - ] - }, - { - "cell_type": "code", - "execution_count": 51, - "id": "d6b763e4", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } + "text/plain": [ + "shape: (12, 4)\n", + "┌──────────────────────────┬────────────┬───────────┬──────────┐\n", + "│ column ┆ null_count ┆ null_rate ┆ null_pct │\n", + "│ --- ┆ --- ┆ --- ┆ --- │\n", + "│ str ┆ u32 ┆ f64 ┆ f64 │\n", + "╞══════════════════════════╪════════════╪═══════════╪══════════╡\n", + "│ Post Employment Position ┆ 47 ┆ 0.94 ┆ 94.0 │\n", + "│ Termination Year ┆ 46 ┆ 0.92 ┆ 92.0 │\n", + "│ Termination Date ┆ 46 ┆ 0.92 ┆ 92.0 │\n", + "│ Termination Reason ┆ 43 ┆ 0.86 ┆ 86.0 │\n", + "│ Hire Year ┆ 38 ┆ 0.76 ┆ 76.0 │\n", + "│ … ┆ … ┆ … ┆ … │\n", + "│ Nick Name ┆ 29 ┆ 0.58 ┆ 58.0 │\n", + "│ Fur Color ┆ 17 ┆ 0.34 ┆ 34.0 │\n", + "│ Prior Position ┆ 9 ┆ 0.18 ┆ 18.0 │\n", + "│ Prior Employer ┆ 9 ┆ 0.18 ┆ 18.0 │\n", + "│ Breed ┆ 1 ┆ 0.02 ┆ 2.0 │\n", + "└──────────────────────────┴────────────┴───────────┴──────────┘" + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "n = mascots.height\n", + "\n", + "mascots_missingness = (\n", + " mascots\n", + " .null_count()\n", + " .transpose(include_header=True, header_name='column', column_names=['null_count'])\n", + " .with_columns([\n", + " (pl.col('null_count') / pl.lit(n)).alias('null_rate'),\n", + " (pl.col('null_count') / pl.lit(n) * 100).alias('null_pct'),\n", + " ])\n", + " .sort('null_count', descending=True)\n", + ")\n", + "\n", + "mascots_missingness.head(12)" + ] + }, + { + "cell_type": "markdown", + "id": "b4a973a0", + "metadata": {}, + "source": [ + "### Quick numeric summary\n", + "\n", + "This is triage, not a final report.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "5bf931c8", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (9, 23)
statisticNameSpeciesBreedPrior PositionPrior EmployerInterview DateHire YearHire DateTermination YearTermination DateTermination ReasonPost Employment PositionNick NameTempermentFur ColorAgeHeightWeightOffer SentStarting SalaryBiterHired
strstrstrstrstrstrstrf64strf64strstrstrstrstrstrf64f64strf64f64f64f64
"count""50""50""49""41""41""50"12.0"12"4.0"4""7""3""21""49""33"50.050.0"50"50.017.050.050.0
"null_count""0""0""1""9""9""0"38.0"38"46.0"46""43""47""29""1""17"0.00.0"0"0.033.00.00.0
"mean"nullnullnullnullnullnull1964.25null1973.75nullnullnullnullnullnull10.32419.284null0.349862.7647060.080.24
"std"nullnullnullnullnullnull43.587164null32.356092nullnullnullnullnullnull1.91626316.461679null0.478518356.5078070.2740480.431419
"min""Ash""Alpaca""African Grey Parrot"" Demolition Expert""ACME Food Storage""1899-02-17"1899.0"1899-02-23"1946.0"1940-05-17""Escaped""Art Thief""Boo Boo""bewildered""black"5.01.25"0.0003"0.09313.00.00.0
"25%"nullnullnullnullnullnull1940.0null1959.0nullnullnullnullnullnull9.17.59null0.09572.00.00.0
"50%"nullnullnullnullnullnull1966.0null1970.0nullnullnullnullnullnull10.313.25null0.09904.00.00.0
"75%"nullnullnullnullnullnull2010.0null1970.0nullnullnullnullnullnull11.626.2null1.010086.00.00.0
"max""Willow""Wolf""Wolf/German Shepard""Zoo""Walmart""2024-09-09"2021.0"2021-11-17"2020.0"2020-11-09""Retired""Traveling Zoo""Yoda""sweet""white"14.777.34"90.0"1.010478.01.01.0
" ], - "source": [ - "celebs_typed = celebs.with_columns([\n", - " pl.col('GPA').cast(pl.Float64),\n", - " pl.col('Workstudy Hourly Rate').cast(pl.Float64),\n", - " pl.col('Loved Library?').cast(pl.Int64),\n", - "])\n", - "\n", - "\n", - "gpa = celebs_typed['GPA'].drop_nulls().to_list()\n", - "plt.figure(figsize=(6,4))\n", - "plt.hist(gpa, bins=12)\n", - "plt.xlabel('GPA')\n", - "plt.ylabel('Count')\n", - "plt.title('GPA distribution')\n", - "plt.tight_layout()\n", - "plt.show()" - ] - }, - { - "cell_type": "markdown", - "id": "07579e3e", - "metadata": {}, - "source": [ - "### Try it yourself (univariate)\n", - "\n", - "Make a bar chart of **counts by College** in the celebs dataset.\n", - "\n", - "- **Minimum**: produce the `college_counts` table.\n", - "- **Stretch**: make the bar chart; optionally plot only the top 10.\n", - "\n", - "
\n", - "Solution\n", - "\n", - "```python\n", - "college_counts = celebs.group_by('College').len().sort('len', descending=True)\n", - "\n", - "top_n = 10\n", - "to_plot = college_counts.head(top_n)\n", - "\n", - "plt.figure(figsize=(8,4))\n", - "plt.bar(to_plot['College'].to_list(), to_plot['len'].to_list())\n", - "plt.xticks(rotation=45, ha='right')\n", - "plt.ylabel('Count')\n", - "plt.title(f'Celebs by College (top {top_n})')\n", - "plt.tight_layout()\n", - "plt.show()\n", - "```\n", - "\n", - "
\n" - ] - }, - { - "cell_type": "code", - "execution_count": 52, - "id": "4c70789d", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } + "text/plain": [ + "shape: (9, 23)\n", + "┌────────────┬────────┬─────────┬─────────────┬───┬────────────┬─────────────┬──────────┬──────────┐\n", + "│ statistic ┆ Name ┆ Species ┆ Breed ┆ … ┆ Offer Sent ┆ Starting ┆ Biter ┆ Hired │\n", + "│ --- ┆ --- ┆ --- ┆ --- ┆ ┆ --- ┆ Salary ┆ --- ┆ --- │\n", + "│ str ┆ str ┆ str ┆ str ┆ ┆ f64 ┆ --- ┆ f64 ┆ f64 │\n", + "│ ┆ ┆ ┆ ┆ ┆ ┆ f64 ┆ ┆ │\n", + "╞════════════╪════════╪═════════╪═════════════╪═══╪════════════╪═════════════╪══════════╪══════════╡\n", + "│ count ┆ 50 ┆ 50 ┆ 49 ┆ … ┆ 50.0 ┆ 17.0 ┆ 50.0 ┆ 50.0 │\n", + "│ null_count ┆ 0 ┆ 0 ┆ 1 ┆ … ┆ 0.0 ┆ 33.0 ┆ 0.0 ┆ 0.0 │\n", + "│ mean ┆ null ┆ null ┆ null ┆ … ┆ 0.34 ┆ 9862.764706 ┆ 0.08 ┆ 0.24 │\n", + "│ std ┆ null ┆ null ┆ null ┆ … ┆ 0.478518 ┆ 356.507807 ┆ 0.274048 ┆ 0.431419 │\n", + "│ min ┆ Ash ┆ Alpaca ┆ African ┆ … ┆ 0.0 ┆ 9313.0 ┆ 0.0 ┆ 0.0 │\n", + "│ ┆ ┆ ┆ Grey Parrot ┆ ┆ ┆ ┆ ┆ │\n", + "│ 25% ┆ null ┆ null ┆ null ┆ … ┆ 0.0 ┆ 9572.0 ┆ 0.0 ┆ 0.0 │\n", + "│ 50% ┆ null ┆ null ┆ null ┆ … ┆ 0.0 ┆ 9904.0 ┆ 0.0 ┆ 0.0 │\n", + "│ 75% ┆ null ┆ null ┆ null ┆ … ┆ 1.0 ┆ 10086.0 ┆ 0.0 ┆ 0.0 │\n", + "│ max ┆ Willow ┆ Wolf ┆ Wolf/German ┆ … ┆ 1.0 ┆ 10478.0 ┆ 1.0 ┆ 1.0 │\n", + "│ ┆ ┆ ┆ Shepard ┆ ┆ ┆ ┆ ┆ │\n", + "└────────────┴────────┴─────────┴─────────────┴───┴────────────┴─────────────┴──────────┴──────────┘" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "mascots.describe()" + ] + }, + { + "cell_type": "markdown", + "id": "26e4a94f", + "metadata": {}, + "source": [ + "### Categorical “shape” (unique counts)\n", + "\n", + "This is a fast way to find columns with many categories (IDs, free text, etc.).\n" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "e6ccf641", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (22, 2)
columnn_unique
stru32
"Name"50
"Interview Date"50
"Weight"50
"Height"49
"Breed"45
……
"Termination Date"5
"Post Employment Position"4
"Offer Sent"2
"Biter"2
"Hired"2
" ], - "source": [ - "college_counts = celebs.group_by('College').len().sort('len', descending=True)\n", - "\n", - "\n", - "top_n = 10\n", - "to_plot = college_counts.head(top_n)\n", - "\n", - "plt.figure(figsize=(8,4))\n", - "plt.bar(to_plot['College'].to_list(), to_plot['len'].to_list())\n", - "plt.xticks(rotation=45, ha='right')\n", - "plt.ylabel('Count')\n", - "plt.title(f'Celebs by College (top {top_n})')\n", - "plt.tight_layout()\n", - "plt.show()" - ] - }, - { - "cell_type": "markdown", - "id": "6d33abb1", - "metadata": {}, - "source": [ - "## 5. Cleaning + feature creation (Penguins)\n", - "\n", - "Cleaning is easiest when it’s tied to a question.\n", - "\n", - "Here we’ll:\n", - "- standardize the `sex` column\n", - "- drop rows missing key measurements (for a specific analysis)\n", - "- create a feature (`bill_ratio`) **before** we summarize\n", - "\n", - "Note: dropping rows is not always appropriate; it depends on your goal.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 53, - "id": "1617731b", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (7, 4)
columnnull_countnull_ratenull_pct
stru32f64f64
"bill_length_mm"20.0058140.581395
"bill_depth_mm"20.0058140.581395
"flipper_length_mm"20.0058140.581395
"body_mass_g"20.0058140.581395
"species"00.00.0
"island"00.00.0
"sex"00.00.0
" - ], - "text/plain": [ - "shape: (7, 4)\n", - "┌───────────────────┬────────────┬───────────┬──────────┐\n", - "│ column ┆ null_count ┆ null_rate ┆ null_pct │\n", - "│ --- ┆ --- ┆ --- ┆ --- │\n", - "│ str ┆ u32 ┆ f64 ┆ f64 │\n", - "╞═══════════════════╪════════════╪═══════════╪══════════╡\n", - "│ bill_length_mm ┆ 2 ┆ 0.005814 ┆ 0.581395 │\n", - "│ bill_depth_mm ┆ 2 ┆ 0.005814 ┆ 0.581395 │\n", - "│ flipper_length_mm ┆ 2 ┆ 0.005814 ┆ 0.581395 │\n", - "│ body_mass_g ┆ 2 ┆ 0.005814 ┆ 0.581395 │\n", - "│ species ┆ 0 ┆ 0.0 ┆ 0.0 │\n", - "│ island ┆ 0 ┆ 0.0 ┆ 0.0 │\n", - "│ sex ┆ 0 ┆ 0.0 ┆ 0.0 │\n", - "└───────────────────┴────────────┴───────────┴──────────┘" - ] - }, - "execution_count": 53, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (22, 2)\n", + "┌──────────────────────────┬──────────┐\n", + "│ column ┆ n_unique │\n", + "│ --- ┆ --- │\n", + "│ str ┆ u32 │\n", + "╞══════════════════════════╪══════════╡\n", + "│ Name ┆ 50 │\n", + "│ Interview Date ┆ 50 │\n", + "│ Weight ┆ 50 │\n", + "│ Height ┆ 49 │\n", + "│ Breed ┆ 45 │\n", + "│ … ┆ … │\n", + "│ Termination Date ┆ 5 │\n", + "│ Post Employment Position ┆ 4 │\n", + "│ Offer Sent ┆ 2 │\n", + "│ Biter ┆ 2 │\n", + "│ Hired ┆ 2 │\n", + "└──────────────────────────┴──────────┘" + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "(\n", + " mascots\n", + " .select(pl.all().n_unique())\n", + " .transpose(include_header=True, header_name='column', column_names=['n_unique'])\n", + " .sort('n_unique', descending=True)\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "11ab2d30", + "metadata": {}, + "source": [ + "### Duplicates (quick check)\n", + "\n", + "First ask: **do we expect duplicates?** If not, check for them.\n", + "\n", + "Note: “duplicate” depends on *which columns* define identity.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "06ecb5d7", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (1, 2)
n_rowsn_duplicate_rows
u32u32
500
" ], - "source": [ - "penguins = (\n", - " penguins_raw\n", - " .with_columns(\n", - " pl.col('sex')\n", - " .str.to_lowercase()\n", - " .str.strip_chars()\n", - " .alias('sex')\n", - " )\n", - " .with_columns(\n", - " pl.when(pl.col('sex').is_null())\n", - " .then(pl.lit('unknown'))\n", - " .otherwise(pl.col('sex'))\n", - " .alias('sex')\n", - " )\n", - ")\n", - "\n", - "(\n", - " penguins\n", - " .null_count()\n", - " .transpose(include_header=True, header_name='column', column_names=['null_count'])\n", - " .with_columns([\n", - " (pl.col('null_count') / pl.lit(penguins.height)).alias('null_rate'),\n", - " (pl.col('null_count') / pl.lit(penguins.height) * 100).alias('null_pct'),\n", - " ])\n", - " .sort('null_count', descending=True)\n", - " .head(12)\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": 54, - "id": "5d2fe4c9", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (5, 8)
speciesislandbill_length_mmbill_depth_mmflipper_length_mmbody_mass_gsexbill_ratio
strstrf64f64i64i64strf64
"Adelie""Torgersen"39.118.71813750"male"2.090909
"Adelie""Torgersen"39.517.41863800"female"2.270115
"Adelie""Torgersen"40.318.01953250"female"2.238889
"Adelie""Torgersen"36.719.31933450"female"1.901554
"Adelie""Torgersen"39.320.61903650"male"1.907767
" - ], - "text/plain": [ - "shape: (5, 8)\n", - "┌─────────┬───────────┬─────────────┬─────────────┬─────────────┬────────────┬────────┬────────────┐\n", - "│ species ┆ island ┆ bill_length ┆ bill_depth_ ┆ flipper_len ┆ body_mass_ ┆ sex ┆ bill_ratio │\n", - "│ --- ┆ --- ┆ _mm ┆ mm ┆ gth_mm ┆ g ┆ --- ┆ --- │\n", - "│ str ┆ str ┆ --- ┆ --- ┆ --- ┆ --- ┆ str ┆ f64 │\n", - "│ ┆ ┆ f64 ┆ f64 ┆ i64 ┆ i64 ┆ ┆ │\n", - "╞═════════╪═══════════╪═════════════╪═════════════╪═════════════╪════════════╪════════╪════════════╡\n", - "│ Adelie ┆ Torgersen ┆ 39.1 ┆ 18.7 ┆ 181 ┆ 3750 ┆ male ┆ 2.090909 │\n", - "│ Adelie ┆ Torgersen ┆ 39.5 ┆ 17.4 ┆ 186 ┆ 3800 ┆ female ┆ 2.270115 │\n", - "│ Adelie ┆ Torgersen ┆ 40.3 ┆ 18.0 ┆ 195 ┆ 3250 ┆ female ┆ 2.238889 │\n", - "│ Adelie ┆ Torgersen ┆ 36.7 ┆ 19.3 ┆ 193 ┆ 3450 ┆ female ┆ 1.901554 │\n", - "│ Adelie ┆ Torgersen ┆ 39.3 ┆ 20.6 ┆ 190 ┆ 3650 ┆ male ┆ 1.907767 │\n", - "└─────────┴───────────┴─────────────┴─────────────┴─────────────┴────────────┴────────┴────────────┘" - ] - }, - "execution_count": 54, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (1, 2)\n", + "┌────────┬──────────────────┐\n", + "│ n_rows ┆ n_duplicate_rows │\n", + "│ --- ┆ --- │\n", + "│ u32 ┆ u32 │\n", + "╞════════╪══════════════════╡\n", + "│ 50 ┆ 0 │\n", + "└────────┴──────────────────┘" + ] + }, + "execution_count": 13, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "mascots.select([\n", + " pl.len().alias('n_rows'),\n", + " pl.struct(pl.all()).is_duplicated().sum().alias('n_duplicate_rows')\n", + "])" + ] + }, + { + "cell_type": "markdown", + "id": "0a60b4cf", + "metadata": {}, + "source": [ + "### Try it yourself (profiling)\n", + "\n", + "- **Minimum**: show the 10 columns with the highest `null_count` in `mascots`.\n", + "- **Stretch**: show the 10 columns with the highest `null_pct`.\n", + "\n", + "Expected result shape: a table with columns like `column`, `null_count`, `null_rate`, `null_pct`.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "b0c6cf5d", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (10, 4)
columnnull_countnull_ratenull_pct
stru32f64f64
"Post Employment Position"470.9494.0
"Termination Year"460.9292.0
"Termination Date"460.9292.0
"Termination Reason"430.8686.0
"Hire Year"380.7676.0
"Hire Date"380.7676.0
"Starting Salary"330.6666.0
"Nick Name"290.5858.0
"Fur Color"170.3434.0
"Prior Position"90.1818.0
" ], - "source": [ - "penguins_clean = penguins.drop_nulls(\n", - " subset=['bill_length_mm','bill_depth_mm','flipper_length_mm','body_mass_g']\n", - ").with_columns(\n", - " (pl.col('bill_length_mm') / pl.col('bill_depth_mm')).alias('bill_ratio')\n", - ")\n", - "\n", - "penguins_clean.head(5)" - ] - }, - { - "cell_type": "markdown", - "id": "38ce381b", - "metadata": {}, - "source": [ - "### Sanity check distributions (univariate, after cleaning)\n", - "\n", - "A quick plot helps confirm we didn’t do something surprising.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 55, - "id": "7b5115d0", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } + "text/plain": [ + "shape: (10, 4)\n", + "┌──────────────────────────┬────────────┬───────────┬──────────┐\n", + "│ column ┆ null_count ┆ null_rate ┆ null_pct │\n", + "│ --- ┆ --- ┆ --- ┆ --- │\n", + "│ str ┆ u32 ┆ f64 ┆ f64 │\n", + "╞══════════════════════════╪════════════╪═══════════╪══════════╡\n", + "│ Post Employment Position ┆ 47 ┆ 0.94 ┆ 94.0 │\n", + "│ Termination Year ┆ 46 ┆ 0.92 ┆ 92.0 │\n", + "│ Termination Date ┆ 46 ┆ 0.92 ┆ 92.0 │\n", + "│ Termination Reason ┆ 43 ┆ 0.86 ┆ 86.0 │\n", + "│ Hire Year ┆ 38 ┆ 0.76 ┆ 76.0 │\n", + "│ Hire Date ┆ 38 ┆ 0.76 ┆ 76.0 │\n", + "│ Starting Salary ┆ 33 ┆ 0.66 ┆ 66.0 │\n", + "│ Nick Name ┆ 29 ┆ 0.58 ┆ 58.0 │\n", + "│ Fur Color ┆ 17 ┆ 0.34 ┆ 34.0 │\n", + "│ Prior Position ┆ 9 ┆ 0.18 ┆ 18.0 │\n", + "└──────────────────────────┴────────────┴───────────┴──────────┘" + ] + }, + "execution_count": 14, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "(\n", + " mascots\n", + " .null_count()\n", + " .transpose(include_header=True, header_name='column', column_names=['null_count'])\n", + " .with_columns([\n", + " (pl.col('null_count') / pl.lit(mascots.height)).alias('null_rate'),\n", + " (pl.col('null_count') / pl.lit(mascots.height) * 100).alias('null_pct'),\n", + " ])\n", + " .sort('null_count', descending=True)\n", + " .head(10)\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "6012baf7", + "metadata": {}, + "source": [ + "## Checkpoint B\n", + "\n", + "You should now have a reusable profiling checklist: shape, schema, preview, missingness (counts/rates), `describe`, `n_unique`, and “do we expect duplicates?”.\n" + ] + }, + { + "cell_type": "markdown", + "id": "621f6939", + "metadata": {}, + "source": [ + "## 4. Univariate exploration\n", + "\n", + "Univariate exploration answers: **“What does this variable look like by itself?”**\n", + "\n", + "- Numeric: distribution (histogram), outliers (boxplot)\n", + "- Categorical: counts (bar chart)\n", + "\n", + "We’ll do a quick categorical count on `mascots` and a histogram on `celebs` GPA.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "c3f7e618", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (23, 2)
Specieslen
stru32
"Dog"10
"Cat"5
"Bird"4
"Rabbit"3
"Horse"3
……
"Hedgehog"1
"Mouse"1
"Duck"1
"Chicken"1
"Snake"1
" ], - "source": [ - "mass = penguins_clean['body_mass_g'].to_list()\n", - "plt.figure(figsize=(6,4))\n", - "plt.hist(mass, bins=15)\n", - "plt.xlabel('Body mass (g)')\n", - "plt.ylabel('Count')\n", - "plt.title('Penguins: body mass after dropping missing values')\n", - "plt.tight_layout()\n", - "plt.show()" - ] - }, - { - "cell_type": "markdown", - "id": "2fa80751", - "metadata": {}, - "source": [ - "## 6. Bivariate exploration\n", - "\n", - "Bivariate exploration asks: **“How do two variables relate?”**\n", - "\n", - "Common patterns:\n", - "- numeric ↔ numeric: scatter plot, correlation\n", - "- category ↔ numeric: boxplot, group summaries\n" - ] - }, - { - "cell_type": "code", - "execution_count": 56, - "id": "6778c7b9", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } + "text/plain": [ + "shape: (23, 2)\n", + "┌──────────┬─────┐\n", + "│ Species ┆ len │\n", + "│ --- ┆ --- │\n", + "│ str ┆ u32 │\n", + "╞══════════╪═════╡\n", + "│ Dog ┆ 10 │\n", + "│ Cat ┆ 5 │\n", + "│ Bird ┆ 4 │\n", + "│ Rabbit ┆ 3 │\n", + "│ Horse ┆ 3 │\n", + "│ … ┆ … │\n", + "│ Hedgehog ┆ 1 │\n", + "│ Mouse ┆ 1 │\n", + "│ Duck ┆ 1 │\n", + "│ Chicken ┆ 1 │\n", + "│ Snake ┆ 1 │\n", + "└──────────┴─────┘" + ] + }, + "execution_count": 15, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "species_counts = (\n", + " mascots\n", + " .group_by('Species')\n", + " .len()\n", + " .sort('len', descending=True)\n", + ")\n", + "species_counts" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "43825124", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "top_n = 10\n", + "top_species = species_counts.head(top_n)\n", + "\n", + "plt.figure(figsize=(8, 4))\n", + "plt.bar(top_species['Species'].to_list(), top_species['len'].to_list())\n", + "plt.xticks(rotation=45, ha='right')\n", + "plt.ylabel('Count')\n", + "plt.title(f'Mascots by Species (top {top_n})')\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "d6b763e4", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsQAAAHUCAYAAAA9az6FAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjYsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvq6yFwwAAAAlwSFlzAAASdAAAEnQB3mYfeAAAOaJJREFUeJzt3Xl0VPXdx/HPZIHERJYsI4YtCMoSpQgaQZQlCgKCxIUiVgXxoCwiwaIFtZIgWtpHBLQEBSy2RVABkYpC2QJUEQZBK2GRRhJAtrCFkDQgmPv84ZM8DpmETJLJnZv7fp2Tk9Pfvb8738m3d+bD9Td3HIZhGAIAAABsKsDsAgAAAAAzEYgBAABgawRiAAAA2BqBGAAAALZGIAYAAICtEYgBAABgawRiAAAA2BqBGAAAALZGIAYAAICtEYgBAABgawRiAAAA2BqBGAD8xJAhQ+RwOJSVlVU8lpWVJYfDoSFDhphWl8PhULdu3dzGkpOT5XA4tH79elNqkvzjbwOgZiAQA7C0vXv36plnnlH79u0VERGh4OBgRURE6JZbbtG4ceO0bdu2EnOKwtwvf0JDQ3Xddddp1KhR+uGHHzw+1hdffFG8/+zZs3391KrE+vXr5XA4lJycbHYpFeIpjANAVQsyuwAAqAjDMDRp0iRNmjRJhYWFat++vQYOHKiIiAidPXtW3377rd58801NnTpVf/7znzVq1KgSx+jatWtx2Dpx4oRWrVql1NRUffjhh9q8ebOaN2/utn9RCC4KxE888YTPn2fDhg21e/du1a1b1+ePVZrdu3friiuuMO3xS+MPfxsANQOBGIAlTZo0ScnJyWrcuLEWLlyozp07l9gnOztb06dP15kzZzweo1u3bm5XTi9cuKDevXtr7dq1mjx5subNm1e8LScnR4sWLdK1116rG264QR999JG2b9+u9u3bV/lz+6Xg4GC1atXKp49xOWY/fmn84W8DoGZgyQQAy9m3b58mT56sWrVqacWKFR7DsCQ5nU69+uqreu6558p13ODg4OKrvi6Xy23b/PnzVVBQoCFDhhSvWZ0zZ06F6l+zZo1uv/12hYWFKSIiQomJidqzZ4/HfUtbJ3vs2DGNGzdOLVu2VFhYmOrVq6eWLVtqyJAh2rdvn6Sf1yR3795dkpSSkuK2RKRo7e+7774rh8Ohd999VytXrlS3bt1Ut25dORyO4se63LKFv/71r7rxxhsVGhoqp9OpoUOH6ujRoyX2i42NVWxsrMdjXLomuaguSdqwYYNb7UX/iClrDfGRI0c0atQoxcbGqlatWoqOjtZ9993ncQnNL/8GaWlp6tatm6688krVqVNHd999t3bv3l3qcwdQM3CFGIDlzJs3TxcvXtRDDz2kuLi4y+4fFFT+lzrDMCTJLRBKP4ffgIAAPfroo2rQoIGuuuoqLViwQK+99prCwsLKffzFixdr4MCBqlWrlgYOHKirr75an3/+uTp16qS2bduW6xj//e9/1blzZ33//ffq0aOH+vXrJ8MwtH//fi1btkwPPPCArrnmGiUmJkr6ObD+cnmIpBLBdPHixVq5cqV69+6t4cOHu32wryzTpk3TqlWrNHDgQPXq1Uuff/655s2bp/Xr12vLli2Kjo4u13Eu1a5dO02cOFEpKSlq2rSpW+i93JrizMxM3XbbbTp8+LASEhI0aNAgHTx4UIsWLdKnn36qJUuWqG/fviXmLV++XMuWLSv+G+zatUufffaZtm7dql27dikqKqpCzwWABRgAYDHdu3c3JBlz586t0PyJEycakoyJEye6jV+4cMFISEgwJBlDhw4tHv/yyy8NSUbPnj2Lx5555hmvazh79qwRERFhBAUFGVu3bnXblpSUZEgyJBmZmZnF45mZmYYkY/DgwcVj//jHPwxJRlJSUonHOH/+vJGbm1v8v9PS0jw+1yLz5s0zJBkOh8NYsWKFx30kGV27dnUbK/obBgcHG9u3b/f4XH75NzQMw2jatKnRtGlTj49RdLy0tLTLPnYRT38bwzCMnj17GpKMyZMnu41/8cUXRmBgoBEREWGcPXu2eLzobxAYGGisWbPGbc748eMNScYf//hHjzUAqBlYMgHAcor+c3zDhg1LbMvKylJycrLbz/Tp0z0eZ/369cX7jB49Wm3atNG6desUFRWlF154oXi/og/T/fIq5WOPPSbJu2UTy5Yt06lTp/TQQw/ppptuctuWnJzs9YfDQkNDS4zVqlVLV155pVfHkaT+/furV69eXs975JFHdOONN7qNFT2XBQsW6Pz5814fszJ++OEHrVq1Sk2aNCmxVObWW2/VoEGDdOrUKX300Ucl5j744IO644473MZKW0IDoGZhyQQAyzFKWdYg/RyIU1JS3MaaNm2qpKSkEvtu2LBBGzZskPRzkGzcuLGGDx+u559/Xo0bN5Yk5ebm6sMPP1TdunV17733Fs+9/vrr1b59e23ZskXffvttuZY7bN++XdLPd7e4VN26ddWuXbviesrStWtXNWzYUFOmTNH27dvVp08fde7cWe3atVNgYOBl53sSHx9foXmXey67d+9Wu3btKnTsivj6668lSbfffruCg4NLbE9ISND8+fP19ddf69FHH3Xbduk/UiQV///g9OnTPqgWgL/gCjEAy7n66qslSYcOHSqxrVu3bjIMQ4Zh6MKFC2UeZ+LEicX7nj9/XhkZGZo1a1ZxCJKk9957T/n5+XrwwQcVEhLiNr/oKnF570lcdLeLq666yuP2Bg0alOs4derU0ebNm/XYY49p27ZtGjNmjG666SY1aNBAEydOvOzzrsxjX+pyz6W0O3z4StHjFf1/5FJF4zk5OSW21atXr8RY0frzn376qWoKBOCXCMQALKforhJr1671+WMVLYl4++23S3yZx+jRoyX9/x0oLqdoScSxY8c8bvd0Z4bSNGrUSO+8846ys7OVnp6uN954Q5GRkcX3ZvaWp6vt5XG55/LLZSABAQG6ePGix/09BdSKKHq80v6WR44cKVEXABCIAVjOkCFDFBQUpMWLF/v0llhfffWVvv76a8XExOjxxx/3+NO2bVudOXNGH3744WWPV3TPYk/LIs6cOaNvvvnG6xodDofi4uI0evRorV69WpL08ccfF28vWkLhqyucZT2XkJAQtW7duni8fv36OnbsmMcr2F999ZXH4wcEBHhVe9F65s8//9xj+E5LS5Mkn98/GoC1EIgBWE7z5s314osv6scff1Tv3r21adMmj/tV9qpj0VKIMWPGaO7cuR5/Xn/9dbd9y9K/f3/Vr19fCxYsKBEAk5OTy728ID093eNt0Yqu1v7yW+UiIyMlSQcOHCjXsb3197//vXjdbpGi5zJo0CDVrl27eDw+Pl4XL150+8IT6ef7AH/xxRcejx8ZGamDBw+Wu55GjRqpR48eysrKKvFhyi1btmjBggWqX7++23pwAOBDdQAs6aWXXpJhGHr55ZfVuXNndejQQfHx8YqIiFBOTo6ysrK0Zs0aSVKXLl28Pn5eXp4WLlyooKAgDR48uNT9EhISdM0112jTpk3auXNnmfdFDg8P1+zZszVw4EDdfvvtbvchTk9PV5cuXbRx48bL1rZmzRo988wzuvXWW9WqVSs5nU798MMPWrZsmQICAvTss88W79uyZUs1bNhQ77//vmrVqqUmTZrI4XDokUceUdOmTb37o3jQu3dvde7cWb/+9a+Ln8vnn3+u2NhYTZkyxW3f0aNHa968eRoxYoTWrl2rxo0b69///rc2bdqkvn37avny5SWOf8cdd+j9999Xv3791KFDBwUFBalLly5l9vStt95S586d9eyzz2rVqlW66aabiu9DHBAQoHnz5lXoThwAai6uEAOwpKJvLNu1a5eSkpJ08eJFLViwQH/84x+1YMECHTt2TCNGjNC2bdv0t7/9zevjL1y4UHl5eerXr1+pHxwrquPxxx+XVL6rxA888IBWrlypDh066MMPP9Rbb72liIgIffnll2rWrFm5arvrrruUlJSkc+fOadmyZZo6dao2btyoHj166F//+pceeOCB4n0DAwO1dOlS3Xbbbfrwww81ceJE/f73v1dmZma5Hutyxo4dq9TUVH3zzTeaPn269uzZoyFDhmjTpk1yOp1u+7Zp00Zr1qxR586d9cknn2j27NmqVauWvvzyS3Xo0MHj8WfMmKFBgwbJ5XLp5Zdf1u9//3utW7euzJquueYaffXVVxo+fLi+++47vfbaa1qxYoV69eqlL774Qv3796+S5w6g5nAYRfcvAgAAAGyIK8QAAACwNQIxAAAAbI1ADAAAAFsjEAMAAMDWCMQAAACwNQIxAAAAbI1ADAAAAFsjEAMAAMDW+Orm/5OTk6MNGzaocePGql27ttnlAAAAoALOnz+vgwcPqmvXrqpXr1655hCI/8+GDRuUmJhodhkAAACoAh9//HG5v6qdQPx/GjduLOnnP16LFi1MrSUvL08ul0vx8fEKDw83tRaUDz2zJvpmTfTNeuiZNVm1bxkZGUpMTCzOduVBIP4/RcskWrRoobi4OFNryc3N1dGjR9W6dWvVqVPH1FpQPvTMmuibNdE366Fn1mT1vnmzBJYP1QEAAMDWCMQAAACwNQIxAAAAbI1ADAAAAFsjEAMAAMDWCMQAAACwNQIxAAAAbI1ADAAAAFsjEAMAAMDWCMQAAACwNQIxAAAAbI1ADAAAAFsjEAMAAMDWCMQAAACwNQIxAAAAbC3I7AIAwGpix39qdgkVljXlbrNLAAC/wxViAAAA2BqBGAAAALZGIAYAAICtEYgBAABgawRiAAAA2BqBGAAAALZGIAYAAICtEYgBAABgawRiAAAA2BqBGAAAALZGIAYAAICtEYgBAABgawRiAAAA2BqBGAAAALZGIAYAAICtEYgBAABga34XiM+ePavnnntOPXv2VHR0tBwOh5KTk8ucYxiGunTpIofDoaeeeqp6CgUAAECN4HeB+OTJk5o9e7bOnz+vxMTEcs2ZOXOmMjIyfFsYAAAAaiS/C8RNmzbV6dOntWHDBv3hD3+47P5ZWVmaMGGCZs6cWQ3VAQAAoKYJMruASzkcDq/2f+KJJ9SjRw/de++9PqoIAAAANZnfBWJvzJ07Vy6XS7t27TK7FAAAAFiUZQPxoUOHNG7cOP3pT39STEyMV3Ozs7N1/Phxt7GiNch5eXnKzc2tsjorIj8/3+03/B89s6aK9q1BqOGLcqqF2a9vVYHzzXromTVZtW95eXlez7FsIB4+fLh+9atfadiwYV7PTU1NVUpKisdtLpdLR48erWx5VcLlcpldArxEz6zJ275NaOebOqpDWlqa2SVUGc4366Fn1mS1vh04cMDrOZYMxIsXL9bKlSv1+eef68yZM27bfvzxR+Xk5CgsLEzBwcEe548cOVIDBgxwG8vIyFBiYqLi4+PVunVrn9VeHvn5+XK5XIqPj1dYWJiptaB86Jk1VbRvPadt9GFVuJyoEEPDWhVqzp4AnTjn3edOzLZqbBezSzAFr5HWZNW+7d692+s5lgzE6enpunjxojp27Fhi25w5czRnzhwtXbq01Nu2OZ1OOZ1Oj9vCw8NVp06dqiy3wsLCwvymFpQPPbMmb/t2tMBaIaymOnHOYble2P31gddIa7Ja38LDw72eY8lAPGTIEHXr1q3EePfu3ZWYmKgxY8bo+uuvr/7CAAAAYDl+GYhXrFih/Px8nT17VpK0a9cuLV68WJLUp08fxcbGKjY21uPchg0begzLAAAAgCd+GYhHjBih/fv3F//vRYsWadGiRZKkzMzMUsMwAAAA4C2/DMRZWVkVmmcY1r0VEgAAAMzhd1/dDAAAAFQnAjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsze8C8dmzZ/Xcc8+pZ8+eio6OlsPhUHJysts+P/30k15//XX16tVLjRo10hVXXKHWrVtr/PjxysnJMaVuAAAAWJPfBeKTJ09q9uzZOn/+vBITEz3uU1BQoOTkZDVt2lTTp0/XZ599pmHDhmn27Nnq3LmzCgoKqrdoAAAAWFaQ2QVcqmnTpjp9+rQcDodOnDihuXPnltgnNDRUmZmZioyMLB7r1q2bmjRpogEDBmjJkiV6+OGHq7NsAAAAWJTfBWKHw3HZfQIDA93CcJH4+HhJ0sGDB6u8LgAAANRMfrdkojLWrVsnSYqLizO5EgAAAFiF310hrqhDhw5p/Pjxuummm9S3b98y983Oztbx48fdxjIyMiRJeXl5ys3N9Vmd5ZGfn+/2G/6PnllTRfvWINTwRTkop6gQw+23lZj9/mIWXiOtyap9y8vL83pOjQjEp06dUp8+fWQYhj744AMFBJR94Ts1NVUpKSket7lcLh09etQXZXrN5XKZXQK8RM+sydu+TWjnmzrgnWGtCs0uwWtpaWlml2AqXiOtyWp9O3DggNdzLB+IT58+rR49eujQoUNat26drrnmmsvOGTlypAYMGOA2lpGRocTERMXHx6t169a+Krdc8vPz5XK5FB8fr7CwMFNrQfnQM2uqaN96Ttvow6pwOVEhhoa1KtScPQE6ce7ynzvxJ6vGdjG7BFPwGmlNVu3b7t27vZ5j6UB8+vRp3XnnncrMzNTatWvVtm3bcs1zOp1yOp0et4WHh6tOnTpVWWaFhYWF+U0tKB96Zk3e9u1ogbVCWE114pzDcr2w++sDr5HWZLW+hYeHez3HsoG4KAzv27dPq1ev1o033mh2SQAAALAgvwzEK1asUH5+vs6ePStJ2rVrlxYvXixJ6tOnjxwOh+666y59/fXXmj59ui5evKjNmzcXz4+Ojlbz5s1NqR0AAADW4peBeMSIEdq/f3/x/160aJEWLVokScrMzJQkbd26VZI0ZsyYEvMHDx6sd9991/eFAgAAwPL8MhBnZWVddh/DsN7tdgAAAOB/atQXcwAAAADeIhADAADA1gjEAAAAsDUCMQAAAGyNQAwAAABbIxADAADA1gjEAAAAsDUCMQAAAGyNQAwAAABbIxADAADA1gjEAAAAsDUCMQAAAGyNQAwAAABbIxADAADA1gjEAAAAsDUCMQAAAGyNQAwAAABbCzK7AAD2Ezv+U7NLkCQ1CDU0oZ3Uc9pGHS1wmF0OAMAkXCEGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC25neB+OzZs3ruuefUs2dPRUdHy+FwKDk52eO+27dv15133qnw8HDVq1dP9913n/bt21e9BQMAAMDS/C4Qnzx5UrNnz9b58+eVmJhY6n579uxRt27d9OOPP+rDDz/UX/7yF+3du1e33367jh8/Xn0FAwAAwNKCzC7gUk2bNtXp06flcDh04sQJzZ071+N+L730kmrXrq3ly5erTp06kqQOHTro2muv1WuvvaY//vGP1Vk2AAAALMrvrhA7HA45HI4y97l48aKWL1+u+++/vzgMSz+H6e7du2vp0qW+LhMAAAA1hN8F4vL4/vvvVVBQoLZt25bY1rZtW2VkZOjcuXMmVAYAAACr8bslE+Vx8uRJSVJERESJbRERETIMQ6dPn9bVV1/tcX52dnaJdcYZGRmSpLy8POXm5lZxxd7Jz893+w3/R8+80yDUMLsESVJUiOH2G9Zg5b6Z/f5iFl4jrcmqfcvLy/N6jiUDcZGyllaUtS01NVUpKSket7lcLh09erTStVUFl8tldgnwEj0rnwntzK7A3bBWhWaXgAqwYt/S0tLMLsFUvEZak9X6duDAAa/nWDIQR0ZGSvr/K8W/dOrUKTkcDtWrV6/U+SNHjtSAAQPcxjIyMpSYmKj4+Hi1bt26Suv1Vn5+vlwul+Lj4xUWFmZqLTVZz2kbq+xYUSGGhrUq1Jw9ATpxruw18PAf9M2arNy3VWO7mF2CKXhfsyar9m337t1ez7FkIG7evLlCQ0O1Y8eOEtt27NihFi1aKCQkpNT5TqdTTqfT47bw8HC3D+qZKSwszG9qqYmOFlT9G+mJcw6fHBe+Rd+syYp9s/trOu9r1mS1voWHh3s9x5IfqgsKClK/fv300Ucf6ezZs8XjBw4cUFpamu677z4TqwMAAICV+OUV4hUrVig/P7847O7atUuLFy+WJPXp00dXXHGFUlJSdPPNN6tv374aP368zp07p5deeklRUVH67W9/a2b5AAAAsBC/DMQjRozQ/v37i//3okWLtGjRIklSZmamYmNj1apVK61fv16/+93v9MADDygoKEgJCQl67bXXFB0dbVbpAAAAsBi/DMRZWVnl2q9Dhw5as2aNb4sBAABAjWbJNcQAAABAVSEQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAWyMQAwAAwNYIxAAAALA1AjEAAABsjUAMAAAAW6tUIA4MDJTL5fK4bdu2bQoMDKzM4QEAAACfq1QgNgyj1G2FhYVyOByVOTwAAADgc5VeMlFa6N22bZvq1q1b2cMDAAAAPhXk7YQZM2ZoxowZkn4Ow4mJiapdu7bbPgUFBcrOztYDDzxQNVUCAAAAPuJ1IHY6nYqLi5MkZWVl6ZprrlG9evXc9qldu7ZuuOEGjRkzpkqKBAAAAHzF60A8aNAgDRo0SJLUvXt3zZo1S61atarywgAAAIDq4HUg/qW0tLSqqgMAAAAwRaUCsfTznSa2bt2q/fv3q6CgoMT2Rx99tLIPAQAAAPhMpQLx3r17dc899+g///mPx1uwORwOAjEAAAD8WqUC8ahRo3Tu3Dl98MEHatu2bYm7TQAAAAD+rlKB2OVyac6cOdxeDQAAAJZVqS/mCA8PV506daqqFgAAAKDaVSoQP/bYY1qwYEFV1QIAAABUu0otmbj++uu1cOFC3XPPPerXr58iIyNL7HPfffdV5iFK9fXXXyslJUUul0s5OTlq0qSJHnroIY0bN05XXHGFTx4TAAAANU+lAvFDDz0kScrMzNTy5ctLbHc4HPrpp58q8xAe7dq1S7feeqtatmyp6dOnKyoqShs3btSkSZO0bds2LVu2rMofEwAAADWTJb+YY8GCBTp37pyWLFmi5s2bS5ISEhJ05MgRzZ49W6dPn1b9+vVNqQ0AAADWUqlA3LVr16qqwyvBwcGSpLp167qN16tXTwEBAapVq5YZZQEAAMCCKvWhOrMMHjxY9erV04gRI7Rv3z6dPXtWy5cv19tvv61Ro0YpLCzM7BIBAABgEZW6QpyQkFDmdofDobVr11bmITyKjY3Vl19+qXvvvbd4yYQkPf3005o+ffpl52dnZ+v48eNuYxkZGZKkvLw85ebmVmm93srPz3f7Dd9oEFry2xUrKirEcPsNa6Bv1mTlvpn9/mIW3tesyap9y8vL83pOpQJxYWGhHA6H29iJEyf03Xffyel06rrrrqvM4UuVlZWlfv366aqrrtLixYsVHR2tLVu2aPLkycrLy9M777xT5vzU1FSlpKR43OZyuXT06FFflO01l8tldgk12oR2VX/MYa0Kq/6g8Dn6Zk1W7JtZn73xF7yvWZPV+nbgwAGv5zgMw6jyf2Lv3btX/fv311tvveWTdcYPPvig0tLStG/fPrflEfPmzdPQoUO1fv36Mh+3tCvEiYmJ2rx5s1q3bl3lNXsjPz9fLpdL8fHxLP/woZ7TNlbZsaJCDA1rVag5ewJ04pzj8hPgF+ibNVm5b6vGdjG7BFPwvmZNVu3b7t271bFjR6WnpysuLq5ccyp1hbg01113nZ599lk999xz2rJlS5Uf/5tvvlGbNm1KNOfmm2+WJKWnp5cZiJ1Op5xOp8dt/vTte2FhYX5TS010tKDq30hPnHP45LjwLfpmTVbsm91f03lfsyar9S08PNzrOT77UF1sbKzS09N9cuyYmBjt3LmzxBqRL7/8UpLUqFEjnzwuAAAAah6fXCGWpCVLligmJsYnx05KSlJiYqJ69OihsWPHKioqSps3b9Yf/vAHtWnTRr179/bJ4wIAAKDmqVQgHjp0aImx8+fP69tvv9WuXbv0pz/9qTKHL9U999yjtWvXasqUKRozZozOnDmjxo0b68knn9SECRO4DzEAAADKrVKBeN26dSXuMhESEqLY2FhNmDCh+KudfaF79+7q3r27z44PAAAAe6hUIM7KyqqiMgAAAABzWPKb6gAAAICqUukP1Z06dUrTpk3T2rVrdfLkSUVFRenOO+9UUlKS6tevXxU1AgAAAD5TqSvEhw4dUvv27fXKK6/ozJkzatKkiXJycvTyyy+rffv2Onz4cFXVCQAAAPhEpQLx888/r4KCAm3ZskU7d+7U6tWrtXPnTm3ZskUFBQV6/vnnq6pOAAAAwCcqFYhXrlypyZMnF39DXJGbb75ZkyZN0ooVKypVHAAAAOBrlQrEZ86cUWxsrMdtzZo105kzZypzeAAAAMDnKhWImzVrpk8//dTjthUrVqhZs2aVOTwAAADgc5W6y8Rjjz2m8ePHq7CwUIMHD9bVV1+tI0eOaP78+XrzzTc1ZcqUqqoTAAAA8IlKBeJnn31W33//vf785z9r5syZxeOGYeiJJ57QuHHjKl0gAAAA4EuVCsQOh0Nvv/22nnnmGaWlpenkyZOKjIxUQkKCrrvuuqqqEQAAAPAZr9cQnz59Wvfff7+WL19ePNayZUsNHz5cL7zwgoYPH669e/fq/vvv18mTJ6u0WAAAAKCqeR2I586dq3//+9/q1atXqfv06tVLO3bscFtGAQAAAPgjrwPx+++/r2HDhikoqPTVFkFBQRo2bJj+8Y9/VKo4AAAAwNe8DsR79+7VTTfddNn92rdvr71791aoKAAAAKC6eB2IL168qODg4MvuFxwcrAsXLlSoKAAAAKC6eB2Ir776au3ateuy++3cuVMNGjSoUFEAAABAdfH6tmtdu3ZVamqqHn/88VKvFF+4cEGzZs1S9+7dK10g/FfseM/fUggA8Myur5sNQg1NaCf1nLZRRwscptSQNeVuUx4X1uD1FeKxY8dqz549uvfee3X48OES2w8fPqzExER99913Gjt2bJUUCQAAAPiK11eI27Ztq5kzZ2rkyJFq1qyZOnTooGbNmkmSMjMztW3bNhUWFmrWrFm64YYbqrxgAAAAoCpV6Jvqhg0bpuuvv16vvvqq0tLStHnzZknSFVdcoV69emnChAnq2LFjlRYKAAAA+EKFv7q5U6dO+uSTT1RYWKgTJ05IkqKiohQQ4PUqDAAAAMA0FQ7ERQICAuR0OquiFgAAAKDacTkXAAAAtkYgBgAAgK0RiAEAAGBrBGIAAADYGoEYAAAAtkYgBgAAgK0RiAEAAGBrBGIAAADYGoEYAAAAtkYgBgAAgK0RiAEAAGBrlg7En3/+ufr06aP69esrNDRU1157rV5++WWzywIAAICFBJldQEUtWLBAjzzyiH7961/rb3/7m8LDw/X999/r8OHDZpcGAAAAC7FkID506JCeeOIJPfnkk0pNTS0e7969u4lVAQAAwIosuWRi7ty5ys/P1+9+9zuzSwEAAIDFWTIQb9y4UREREdqzZ4/atWunoKAgOZ1ODR8+XLm5uWaXBwAAAAux7JKJ//73vxowYIAmTJig6dOna+vWrZo4caLS09P1r3/9Sw6Ho9T52dnZOn78uNtYRkaGJCkvL8/0UJ2fn+/22181CDXMLsFvRIUYbr9hDfTNmuib9fhDz8x+b7ciq+SRS+Xl5Xk9x2EYhuVeUa677jr95z//0R/+8AeNHz++eHzGjBlKSkrS6tWrdeedd5Y6Pzk5WSkpKR63vfHGG2rSpEmV1wwAAADfO3DggJ5++mmlp6crLi6uXHMseYU4MjJS//nPf3TXXXe5jffu3VtJSUnavn17mYF45MiRGjBggNtYRkaGEhMTFR8fr9atW/uk7vLKz8+Xy+VSfHy8wsLCTK2lLD2nbTS7BL8RFWJoWKtCzdkToBPnSv+vE/Av9M2a6Jv1+EPPVo3tYsrjWplV8sildu/e7fUcSwbitm3bavPmzSXGiy52BwSUvTTa6XTK6XR63BYeHq46depUvsgqEBYW5je1eHK0gDeiS5045+DvYkH0zZrom/WY2TN/fj/1d/6eRy4VHh7u9RxLfqju/vvvlyStWLHCbfyzzz6TJHXs2LHaawIAAIA1WfIKcc+ePdWvXz9NmjRJhYWF6tixo7766iulpKSob9++uu2228wuEQAAABZhySvEkvTBBx8oKSlJs2fPVu/evTVr1iyNHTtWixcvNrs0AAAAWIglrxBLUmhoqKZMmaIpU6aYXQoAAAAszLJXiAEAAICqQCAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGs1JhDPnTtXDodD4eHhZpcCAAAAC6kRgfjQoUMaN26cYmJizC4FAAAAFlMjAvHw4cPVpUsX9ejRw+xSAAAAYDGWD8Tz58/Xhg0blJqaanYpAAAAsCBLB+Ls7GwlJSVpypQpatSokdnlAAAAwIKCzC6gMkaOHKmWLVtqxIgRXs3Lzs7W8ePH3cYyMjIkSXl5ecrNza2yGisiPz/f7be/ahBqmF2C34gKMdx+wxromzXRN+vxh56Z/d5uRVbJI5fKy8vzeo5lA/GSJUv0ySef6Ouvv5bD4fBqbmpqqlJSUjxuc7lcOnr0aFWUWGkul8vsEso0oZ3ZFfifYa0KzS4BFUDfrIm+WY+ZPUtLSzPtsa3O3/PIpQ4cOOD1HEsG4ry8PI0aNUqjR49WTEyMcnJyJEk//vijJCknJ0fBwcEKCwvzOH/kyJEaMGCA21hGRoYSExMVHx+v1q1b+7T+y8nPz5fL5VJ8fHypz8Ef9Jy20ewS/EZUiKFhrQo1Z0+ATpzz7h9oMA99syb6Zj3+0LNVY7uY8rhWZpU8cqndu3d7PceSgfjEiRM6duyYpk6dqqlTp5bYXr9+ffXv318ff/yxx/lOp1NOp9PjtvDwcNWpU6cqy62wsLAwv6nFk6MFvBFd6sQ5B38XC6Jv1kTfrMfMnvnz+6m/8/c8cqmKfCeFJQNxgwYNPP6njylTpmjDhg1asWKFoqKiTKgMAAAAVmPJQBwSEqJu3bqVGH/33XcVGBjocRsAAADgiaVvuwYAAABUVo0KxO+++26FbrUBAAAA+6pRgRgAAADwFoEYAAAAtkYgBgAAgK0RiAEAAGBrBGIAAADYGoEYAAAAtkYgBgAAgK0RiAEAAGBrBGIAAADYGoEYAAAAtkYgBgAAgK0RiAEAAGBrBGIAAADYGoEYAAAAtkYgBgAAgK0RiAEAAGBrQWYXYGex4z/1ON4g1NCEdlLPaRt1tMBRvUUBAADYDFeIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGsEYgAAANgagRgAAAC2RiAGAACArRGIAQAAYGuWDcTr1q3T0KFD1apVK4WFhalhw4bq37+/tm3bZnZpAAAAsBDLBuJZs2YpKytLY8aM0WeffaYZM2YoOztbHTt21Lp168wuDwAAABYRZHYBFTVz5kw5nU63sV69eqlFixZ69dVXlZCQYFJlAAAAsBLLXiG+NAxLUnh4uNq0aaODBw+aUBEAAACsyLKB2JMzZ85o+/btiouLM7sUAAAAWIRll0x4MmrUKOXn5+uFF14oc7/s7GwdP37cbSwjI0OSlJeXp9zcXJ/V+EsNQg2P41Ehhttv+D96Zk30zZrom/X4Q8+q6729JsnPz3f7bRV5eXlez3EYhlEjXlF+//vfa/LkyXrzzTf11FNPlblvcnKyUlJSPG5744031KRJE1+UCAAAAB87cOCAnn76aaWnp5d71UCNuEKckpKiyZMn65VXXrlsGJakkSNHasCAAW5jGRkZSkxMVHx8vFq3bu2rUt30nLbR43hUiKFhrQo1Z0+ATpxzVEstqBx6Zk30zZrom/X4Q89Wje1iyuNaWX5+vlwul+Lj4xUWFmZ2OeW2e/dur+dYPhCnpKQoOTlZycnJev7558s1x+l0evxQnvTzB/Pq1KlTlSWW6mhB2S8KJ845LrsP/As9syb6Zk30zXrM7Fl1vbfXRGFhYZb6+4WHh3s9x9Ifqnv55ZeVnJysF198URMnTjS7HAAAAFiQZa8QT506VS+99JJ69eqlu+++W5s3b3bb3rFjR5MqAwAAgJVYNhB/8sknkqSVK1dq5cqVJbbXkM8KAgAAwMcsG4jXr19vdgkAAACoASy9hhgAAACoLAIxAAAAbI1ADAAAAFsjEAMAAMDWCMQAAACwNQIxAAAAbI1ADAAAAFsjEAMAAMDWCMQAAACwNQIxAAAAbI1ADAAAAFsjEAMAAMDWCMQAAACwNQIxAAAAbI1ADAAAAFsjEAMAAMDWgswuAAAAwNdix39qdgmW0yDU0IR2Us9pG3W0wFHh42RNubvqivIRrhADAADA1gjEAAAAsDUCMQAAAGyNQAwAAABbIxADAADA1gjEAAAAsDUCMQAAAGyNQAwAAABbIxADAADA1gjEAAAAsDUCMQAAAGyNQAwAAABbIxADAADA1gjEAAAAsDUCMQAAAGyNQAwAAABbIxADAADA1iwbiPPy8pSUlKSYmBiFhISoXbt2ev/9980uCwAAABYTZHYBFXXfffdp69atmjJliq677jotWLBAgwYNUmFhoR566CGzywMAAIBFWDIQf/bZZ1q9enVxCJak7t27a//+/Xr22Wc1cOBABQYGmlwlAAAArMCSSyaWLl2q8PBwDRgwwG38scce0+HDh7VlyxaTKgMAAIDVWDIQp6enq3Xr1goKcr/A3bZt2+LtAAAAQHlYcsnEyZMndc0115QYj4iIKN5eluzsbB0/ftxtbNeuXZKkb7/9Vnl5eVVUadnqnM3yOF77R+nAgULVPhOgOuerpRRUEj2zJvpmTfTNeuiZNVVV36r7v9zv27dPknT+fPmLtmQgliSHw1GhbZKUmpqqlJQUj9v85QN5G80uAF6jZ9ZE36yJvlkPPbOmquhbx9QqOEgFHDx4UO3bty/XvpYMxJGRkR6vAp86dUrS/18pLs3IkSNLrD/Ozc3V3r17dcMNN6h27dpVV2wFZGRkKDExUR9//LFatGhhai0oH3pmTfTNmuib9dAza7Jq386fP6+DBw+qa9eu5Z5jyUB8ww03aOHChbp48aLbOuIdO3ZIkq6//voy5zudTjmdzhLjnTp1qtpCK6lFixaKi4szuwx4gZ5ZE32zJvpmPfTMmqzYt/JeGS5iyQ/V3XvvvcrLy9OSJUvcxv/6178qJiZGt9xyi0mVAQAAwGoseYW4d+/e6tGjh0aMGKHc3Fy1aNFCCxcu1MqVKzV//nzuQQwAAIBys2QglqSPPvpIL7zwgl566SWdOnVKrVq10sKFC/Xggw+aXRoAAAAsxLKBODw8XDNmzNCMGTPMLqXKRUdHa+LEiYqOjja7FJQTPbMm+mZN9M166Jk12alvDsMwDLOLAAAAAMxiyQ/VAQAAAFWFQAwAAABbIxADAADA1gjE1WTdunUaOnSoWrVqpbCwMDVs2FD9+/fXtm3byjU/OztbQ4YMUVRUlK644gp16tRJa9eu9XHV9laZnr377rtyOBwef44ePVoN1dvXN998o7vvvltNmjRRaGioIiIi1KlTJ82fP79c8znXzFGZvnG++Y+5c+fK4XAoPDy8XPtzvpnPm57V5HPNsneZsJpZs2bp5MmTGjNmjNq0aaPjx49r6tSp6tixo/75z38qISGh1Lnnz5/XHXfcoZycHM2YMUNOp1MzZ85Ur169tGbNGq++mhDlV5meFZk3b55atWrlNhYZGemrkiEpJydHjRs31qBBg9SwYUPl5+frvffe0yOPPKKsrCy9+OKLpc7lXDNPZfpWhPPNXIcOHdK4ceMUExOjM2fOXHZ/zjfzeduzIjXyXDNQLY4dO1Zi7OzZs8ZVV11l3HHHHWXOnTlzpiHJ2LRpU/HYhQsXjDZt2hjx8fFVXit+VpmezZs3z5BkbN261VflwUu33HKL0bhx4zL34VzzP+XpG+ebf+jbt6/Rr18/Y/DgwUZYWNhl9+d8M5+3PavJ5xpLJqqJ0+ksMRYeHq42bdro4MGDZc5dunSpWrZsqU6dOhWPBQUF6eGHH5bL5dKhQ4eqvF5UrmfwP1FRUQoKKvs/inGu+Z/y9A3mmz9/vjZs2KDU1NRyz+F8M1dFelaTEYhNdObMGW3fvl1xcXFl7peenq62bduWGC8a27lzp0/qQ0nl7VmRvn37KjAwUBEREbrvvvuUnp7u4wpRpLCwUBcvXtTx48eVmpqqf/7zn/rd735X5hzONfNVpG9FON/MkZ2draSkJE2ZMkWNGjUq9zzON/NUtGdFauK5xj+7TTRq1Cjl5+frhRdeKHO/kydPKiIiosR40djJkyd9Uh9KKm/PGjRooBdeeEEdO3ZUnTp1tGPHDk2ZMkUdO3bUF198oV/96lfVVLF9jRw5Um+//bYkqVatWnrjjTf05JNPljmHc818Fekb55u5Ro4cqZYtW2rEiBFezeN8M09Fe1ajzzWz12zY1YsvvmhIMt58883L7hscHGwMHz68xPimTZsMScbChQt9USIu4U3PPMnMzDTCw8ONe+65p4orgyf79+83tm7danz66afG8OHDjYCAAON//ud/ypzDuWa+ivTNE8636rF48WKjVq1axs6dO4vHyrselfPNHJXpmSc15VzjCrEJUlJSNHnyZL3yyit66qmnLrt/ZGSkx38pnzp1SpI8/gsbVcvbnnkSGxur2267TZs3b67i6uBJkyZN1KRJE0lSnz59JEkTJkzQ4MGDFR0d7XEO55r5KtI3TzjffC8vL0+jRo3S6NGjFRMTo5ycHEnSjz/+KOnnO4cEBwcrLCzM43zOt+pX2Z55UlPONdYQV7OUlBQlJycrOTlZzz//fLnm3HDDDdqxY0eJ8aKx66+/vkprhLuK9Kw0hmEoIIDTzgzx8fG6ePGi9u3bV+o+nGv+pzx9Kw3nm2+dOHFCx44d09SpU1W/fv3in4ULFyo/P1/169fXb37zm1Lnc75Vv8r2rDQ14lwz+Qq1rUyaNMmQZLz44otezUtNTTUkGZs3by4eu3DhghEXF2fccsstVV0mfqGiPfNk3759Rnh4uJGYmFgFlcFbjzzyiBEQEGBkZ2eXug/nmv8pT9884XzzvYKCAiMtLa3Ez1133WWEhIQYaWlpxo4dO0qdz/lW/SrbM09qyrlGIK4mr732miHJ6NWrl/Hll1+W+CkydOhQIzAw0MjKyioeO3funBEXF2c0btzYeO+994zVq1cb9957rxEUFGSsX7/ejKdjC5Xp2R133GGkpKQYS5cuNdauXWtMnz7diImJMa688kqvX2zgnWHDhhm//e1vjQ8++MBYv369sXjxYmPgwIGGJOPZZ58t3o9zzb9Upm+cb/7F03pUzjf/Vt6e1eRzjUBcTbp27WpIKvWnyODBgw1JRmZmptv8o0ePGo8++qgRERFhhISEGB07djRWr15dzc/CXirTs6SkJKNNmzbGlVdeaQQFBRkxMTHGww8/bHz33XcmPBN7+ctf/mLcfvvtRlRUlBEUFGTUq1fP6Nq1q/H3v//dbT/ONf9Smb5xvvkXT+GK882/lbdnNflccxiGYVTL2gwAAADAD1l8BTQAAABQOQRiAAAA2BqBGAAAALZGIAYAAICtEYgBAABgawRiAAAA2BqBGAAAALZGIAYAAICtEYgBAABgawRiALCwb7/9Vo8//riaN2+u0NBQhYaG6tprr9WTTz6pr776qni/5ORkORyO4p9atWqpWbNmGjNmjHJyckoc95lnnpHD4VDfvn2r8dkAgDmCzC4AAFAxb7/9tp566im1bNlSY8aMUVxcnBwOh3bv3q2FCxfq5ptvVkZGhpo3b148Z+XKlapbt67Onj2rzz77TDNmzJDL5dKmTZvkcDgkSRcuXND8+fOL9z906JAaNmxoynMEgOpAIAYAC/riiy80cuRI3X333Vq8eLFq1apVvC0hIUGjRo3SokWLFBoa6javQ4cOioqKkiT16NFDJ0+e1N///ndt2rRJnTt3liQtW7ZMx48f1913361PP/1Uf/3rX/X8889X35MDgGrGkgkAsKBXX31VgYGBevvtt93C8C8NGDBAMTExZR6nY8eOkqT9+/cXj73zzjuqVauW5s2bp8aNG2vevHkyDKPqigcAP0MgBgCL+emnn5SWlqabbrpJV199daWOlZGRIUmKjo6WJP3www9atWqV+vfvr+joaA0ePFgZGRnauHFjpesGAH9FIAYAizlx4oQKCgrUtGnTEtt++uknXbx4sfjn0iu7RdtzcnL03nvv6a233lLjxo11++23S5LmzZunwsJCPf7445KkoUOHyuFw6J133vH9EwMAkxCIAaAG6dChg4KDg4t/pk6d6ra9QYMGCg4OVv369fXwww+rffv2WrlypUJCQmQYRvEyiR49ekiSmjVrpm7dumnJkiXKzc014ykBgM/xoToAsJioqCiFhoa6rfstsmDBAv33v//VkSNHdM8995TYvmbNGtWtW1fBwcFq1KiRIiMji7etW7dOmZmZeuaZZ9zC769//WulpaVp4cKFevLJJ33zpADARARiALCYwMBAJSQkaNWqVTpy5IjbOuI2bdpIkrKysjzO/dWvflV8l4lLFS2LeP311/X666973E4gBlATsWQCACxowoQJ+umnnzR8+HBduHCh0sc7ffq0li5dqs6dOystLa3Ez29+8xtt3bpV6enpVVA9APgXrhADgAV17txZM2fO1OjRo9W+fXs98cQTiouLU0BAgI4cOaIlS5ZIkurUqVOu47333ns6d+6cnn76aXXr1q3E9sjISL333nt65513NG3atKp8KgBgOofBzSUBwLL+/e9/a8aMGVq/fr0OHz4sh8OhRo0a6dZbb9XgwYOVkJAg6eevbk5JSdHx48c9Lpm48cYbdfjwYR08eLDU+xp36tRJGRkZOnToUKn7AIAVEYgBAABga6whBgAAgK0RiAEAAGBrBGIAAADYGoEYAAAAtkYgBgAAgK0RiAEAAGBrBGIAAADYGoEYAAAAtkYgBgAAgK0RiAEAAGBrBGIAAADYGoEYAAAAtkYgBgAAgK0RiAEAAGBr/wuIzyzUrt8tcQAAAABJRU5ErkJggg==", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "celebs_typed = celebs.with_columns([\n", + " pl.col('GPA').cast(pl.Float64),\n", + " pl.col('Workstudy Hourly Rate').cast(pl.Float64),\n", + " pl.col('Loved Library?').cast(pl.Int64),\n", + "])\n", + "\n", + "\n", + "gpa = celebs_typed['GPA'].drop_nulls().to_list()\n", + "plt.figure(figsize=(6,4))\n", + "plt.hist(gpa, bins=12)\n", + "plt.xlabel('GPA')\n", + "plt.ylabel('Count')\n", + "plt.title('GPA distribution')\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "07579e3e", + "metadata": {}, + "source": [ + "### Try it yourself (univariate)\n", + "\n", + "Make a bar chart of **counts by College** in the celebs dataset.\n", + "\n", + "- **Minimum**: produce the `college_counts` table.\n", + "- **Stretch**: make the bar chart; optionally plot only the top 10.\n", + "\n", + "
\n", + "Solution\n", + "\n", + "```python\n", + "college_counts = celebs.group_by('College').len().sort('len', descending=True)\n", + "\n", + "top_n = 10\n", + "to_plot = college_counts.head(top_n)\n", + "\n", + "plt.figure(figsize=(8,4))\n", + "plt.bar(to_plot['College'].to_list(), to_plot['len'].to_list())\n", + "plt.xticks(rotation=45, ha='right')\n", + "plt.ylabel('Count')\n", + "plt.title(f'Celebs by College (top {top_n})')\n", + "plt.tight_layout()\n", + "plt.show()\n", + "```\n", + "\n", + "
\n" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "4c70789d", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "college_counts = celebs.group_by('College').len().sort('len', descending=True)\n", + "\n", + "\n", + "top_n = 10\n", + "to_plot = college_counts.head(top_n)\n", + "\n", + "plt.figure(figsize=(8,4))\n", + "plt.bar(to_plot['College'].to_list(), to_plot['len'].to_list())\n", + "plt.xticks(rotation=45, ha='right')\n", + "plt.ylabel('Count')\n", + "plt.title(f'Celebs by College (top {top_n})')\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "6d33abb1", + "metadata": {}, + "source": [ + "## 5. Cleaning + feature creation (Penguins)\n", + "\n", + "Cleaning is easiest when it’s tied to a question.\n", + "\n", + "Here we’ll:\n", + "- standardize the `sex` column\n", + "- drop rows missing key measurements (for a specific analysis)\n", + "- create a feature (`bill_ratio`) **before** we summarize\n", + "\n", + "Note: dropping rows is not always appropriate; it depends on your goal.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "1617731b", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (7, 4)
columnnull_countnull_ratenull_pct
stru32f64f64
"bill_length_mm"20.0058140.581395
"bill_depth_mm"20.0058140.581395
"flipper_length_mm"20.0058140.581395
"body_mass_g"20.0058140.581395
"species"00.00.0
"island"00.00.0
"sex"00.00.0
" ], - "source": [ - "x = penguins_clean['bill_length_mm'].to_list()\n", - "y = penguins_clean['bill_depth_mm'].to_list()\n", - "\n", - "plt.figure(figsize=(6,4))\n", - "plt.scatter(x, y, s=12, alpha=0.6)\n", - "plt.xlabel('Bill length (mm)')\n", - "plt.ylabel('Bill depth (mm)')\n", - "plt.title('Penguins: bill length vs bill depth')\n", - "plt.tight_layout()\n", - "plt.show()" - ] - }, - { - "cell_type": "code", - "execution_count": 57, - "id": "286f536a", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (1, 2)
corr_bill_len_depthcorr_flipper_mass
f64f64
-0.2350530.871202
" - ], - "text/plain": [ - "shape: (1, 2)\n", - "┌─────────────────────┬───────────────────┐\n", - "│ corr_bill_len_depth ┆ corr_flipper_mass │\n", - "│ --- ┆ --- │\n", - "│ f64 ┆ f64 │\n", - "╞═════════════════════╪═══════════════════╡\n", - "│ -0.235053 ┆ 0.871202 │\n", - "└─────────────────────┴───────────────────┘" - ] - }, - "execution_count": 57, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (7, 4)\n", + "┌───────────────────┬────────────┬───────────┬──────────┐\n", + "│ column ┆ null_count ┆ null_rate ┆ null_pct │\n", + "│ --- ┆ --- ┆ --- ┆ --- │\n", + "│ str ┆ u32 ┆ f64 ┆ f64 │\n", + "╞═══════════════════╪════════════╪═══════════╪══════════╡\n", + "│ bill_length_mm ┆ 2 ┆ 0.005814 ┆ 0.581395 │\n", + "│ bill_depth_mm ┆ 2 ┆ 0.005814 ┆ 0.581395 │\n", + "│ flipper_length_mm ┆ 2 ┆ 0.005814 ┆ 0.581395 │\n", + "│ body_mass_g ┆ 2 ┆ 0.005814 ┆ 0.581395 │\n", + "│ species ┆ 0 ┆ 0.0 ┆ 0.0 │\n", + "│ island ┆ 0 ┆ 0.0 ┆ 0.0 │\n", + "│ sex ┆ 0 ┆ 0.0 ┆ 0.0 │\n", + "└───────────────────┴────────────┴───────────┴──────────┘" + ] + }, + "execution_count": 19, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "penguins = (\n", + " penguins_raw\n", + " .with_columns(\n", + " pl.col('sex')\n", + " .str.to_lowercase()\n", + " .str.strip_chars()\n", + " .alias('sex')\n", + " )\n", + " .with_columns(\n", + " pl.when(pl.col('sex').is_null())\n", + " .then(pl.lit('unknown'))\n", + " .otherwise(pl.col('sex'))\n", + " .alias('sex')\n", + " )\n", + ")\n", + "\n", + "(\n", + " penguins\n", + " .null_count()\n", + " .transpose(include_header=True, header_name='column', column_names=['null_count'])\n", + " .with_columns([\n", + " (pl.col('null_count') / pl.lit(penguins.height)).alias('null_rate'),\n", + " (pl.col('null_count') / pl.lit(penguins.height) * 100).alias('null_pct'),\n", + " ])\n", + " .sort('null_count', descending=True)\n", + " .head(12)\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "5d2fe4c9", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (5, 8)
speciesislandbill_length_mmbill_depth_mmflipper_length_mmbody_mass_gsexbill_ratio
strstrf64f64i64i64strf64
"Adelie""Torgersen"39.118.71813750"male"2.090909
"Adelie""Torgersen"39.517.41863800"female"2.270115
"Adelie""Torgersen"40.318.01953250"female"2.238889
"Adelie""Torgersen"36.719.31933450"female"1.901554
"Adelie""Torgersen"39.320.61903650"male"1.907767
" ], - "source": [ - "penguins_clean.select([\n", - " pl.corr('bill_length_mm', 'bill_depth_mm').alias('corr_bill_len_depth'),\n", - " pl.corr('flipper_length_mm', 'body_mass_g').alias('corr_flipper_mass'),\n", - "])" - ] - }, - { - "cell_type": "code", - "execution_count": 58, - "id": "c50bcf16", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (3, 5)
speciesnmean_massmedian_massp90_mass
stru32f64f64f64
"Gentoo"1235076.016265000.05700.0
"Chinstrap"683733.0882353700.04150.0
"Adelie"1513700.6622523700.04300.0
" - ], - "text/plain": [ - "shape: (3, 5)\n", - "┌───────────┬─────┬─────────────┬─────────────┬──────────┐\n", - "│ species ┆ n ┆ mean_mass ┆ median_mass ┆ p90_mass │\n", - "│ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", - "│ str ┆ u32 ┆ f64 ┆ f64 ┆ f64 │\n", - "╞═══════════╪═════╪═════════════╪═════════════╪══════════╡\n", - "│ Gentoo ┆ 123 ┆ 5076.01626 ┆ 5000.0 ┆ 5700.0 │\n", - "│ Chinstrap ┆ 68 ┆ 3733.088235 ┆ 3700.0 ┆ 4150.0 │\n", - "│ Adelie ┆ 151 ┆ 3700.662252 ┆ 3700.0 ┆ 4300.0 │\n", - "└───────────┴─────┴─────────────┴─────────────┴──────────┘" - ] - }, - "execution_count": 58, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (5, 8)\n", + "┌─────────┬───────────┬─────────────┬─────────────┬─────────────┬────────────┬────────┬────────────┐\n", + "│ species ┆ island ┆ bill_length ┆ bill_depth_ ┆ flipper_len ┆ body_mass_ ┆ sex ┆ bill_ratio │\n", + "│ --- ┆ --- ┆ _mm ┆ mm ┆ gth_mm ┆ g ┆ --- ┆ --- │\n", + "│ str ┆ str ┆ --- ┆ --- ┆ --- ┆ --- ┆ str ┆ f64 │\n", + "│ ┆ ┆ f64 ┆ f64 ┆ i64 ┆ i64 ┆ ┆ │\n", + "╞═════════╪═══════════╪═════════════╪═════════════╪═════════════╪════════════╪════════╪════════════╡\n", + "│ Adelie ┆ Torgersen ┆ 39.1 ┆ 18.7 ┆ 181 ┆ 3750 ┆ male ┆ 2.090909 │\n", + "│ Adelie ┆ Torgersen ┆ 39.5 ┆ 17.4 ┆ 186 ┆ 3800 ┆ female ┆ 2.270115 │\n", + "│ Adelie ┆ Torgersen ┆ 40.3 ┆ 18.0 ┆ 195 ┆ 3250 ┆ female ┆ 2.238889 │\n", + "│ Adelie ┆ Torgersen ┆ 36.7 ┆ 19.3 ┆ 193 ┆ 3450 ┆ female ┆ 1.901554 │\n", + "│ Adelie ┆ Torgersen ┆ 39.3 ┆ 20.6 ┆ 190 ┆ 3650 ┆ male ┆ 1.907767 │\n", + "└─────────┴───────────┴─────────────┴─────────────┴─────────────┴────────────┴────────┴────────────┘" + ] + }, + "execution_count": 20, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "penguins_clean = penguins.drop_nulls(\n", + " subset=['bill_length_mm','bill_depth_mm','flipper_length_mm','body_mass_g']\n", + ").with_columns(\n", + " (pl.col('bill_length_mm') / pl.col('bill_depth_mm')).alias('bill_ratio')\n", + ")\n", + "\n", + "penguins_clean.head(5)" + ] + }, + { + "cell_type": "markdown", + "id": "38ce381b", + "metadata": {}, + "source": [ + "### Sanity check distributions (univariate, after cleaning)\n", + "\n", + "A quick plot helps confirm we didn’t do something surprising.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "7b5115d0", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "mass = penguins_clean['body_mass_g'].to_list()\n", + "plt.figure(figsize=(6,4))\n", + "plt.hist(mass, bins=15)\n", + "plt.xlabel('Body mass (g)')\n", + "plt.ylabel('Count')\n", + "plt.title('Penguins: body mass after dropping missing values')\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "2fa80751", + "metadata": {}, + "source": [ + "## 6. Bivariate exploration\n", + "\n", + "Bivariate exploration asks: **“How do two variables relate?”**\n", + "\n", + "Common patterns:\n", + "- numeric ↔ numeric: scatter plot, correlation\n", + "- category ↔ numeric: boxplot, group summaries\n" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "6778c7b9", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsMAAAHUCAYAAADftyX8AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjYsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvq6yFwwAAAAlwSFlzAAASdAAAEnQB3mYfeAAAlL5JREFUeJzt3Xl8FPX9P/DXzF7Z7GZzbUIIkJAE5BCVoPJFUTk8EBQErLVqW6y21ioeba3WAwH1p7Wtt2KrVDxq633ggdYDULGIcigoKIFwhpD72uy98/sj7DabbPac3Z3dfT0fDx8tm9nd93w+85lMJq99R5AkSQIRERERUQYSk10AEREREVGy8GKYiIiIiDIWL4aJiIiIKGPxYpiIiIiIMhYvhomIiIgoY/FimIiIiIgyFi+GiYiIiChj8WKYiIiIiDIWL4aJiIiIKGPxYpiIiIiIMhYvhomIiIgoY/FimIhkt2bNGgiCgCVLliS7lKCWLFkCQRCwZs2asJ8zdepUCILg99hA+xto22CGDx+O4cOHh729kl166aUQBAF79uxJdilRiXQu9uzZA0EQcOmll/o9HmgcBto2WoIgYOrUqbK8lpyiWV9EycCLYaI4EwTB7z+VSgWz2Yzp06fj+eefT3Z5RFHhhQ49/fTTEAQBTz/9dLJLIYqJOtkFEGWKxYsXAwCcTie+//57vPHGG1i9ejU2btyI+++/P8nVyWvixInYvn07zGZzskuR3bPPPovu7u5kl0EKM2TIEGzfvh25ubnJLoWIIsSLYaIE6fsr9I8++ghnnnkmHnzwQVx77bVp8+txAMjOzsbo0aOTXUZclJWVJbsEUiCNRpO2xzxRumNMgihJTj/9dIwePRqSJOHLL7/0PX7gwAEsXLgQlZWV0Ol0KCwsxJw5c/y28er9q+pXXnkFEydORHZ2NgoKCvCTn/wEBw8eDPjeX375Jc466yzk5OTAZDLhjDPOwH//+9+Av/oOlW+MJkPrcrlw9913Y+TIkdDpdBg2bBhuuukmOByOfq//6aefYvbs2Rg6dCh0Oh1KSkowadIkLF26dICRjc4zzzyD6upq6PV6FBcX47LLLkN9fX1Y+5sI//73vzFt2jTk5+cjKysLY8aMwV133QW73d5vW2+GtKmpCVdccQUGDx4MnU6Ho48+GitWrAj4+na7HUuWLPEddxUVFbjttttgt9v7ZVKHDx/uG/9p06b5xYAC+fvf/45jjjkGWVlZGDRoEK644gq0t7eHtd+//vWvIQgCVq5cGfDr69evhyAIuOCCC3yPHT58GDfccANGjRoFg8GAvLw8jBo1Cpdeeil2794d1vt6tbe3Y+HChRgyZAiysrIwduxYPPzww5AkyW87uXPAfTkcDtx5552oqqrqNz8DcblcWLZsGSZNmgSTyYTs7GxUV1fj0UcfhcfjGbD+HTt2YO7cuSgoKIDBYMApp5yC//znP37bT506Fb/4xS8AAL/4xS/8joFAOfFIzk9EicY7w0RJ5P2G6r2I2LRpE8466yy0tLRgxowZmD9/PpqamvDGG2/glFNOweuvv45Zs2b1e51ly5Zh5cqVmDNnDqZMmYIvvvgCL774Ir7++mts2bIFOp3Ot+2nn36Ks846C06nE+effz6qqqqwdetWTJs2DdOnT0/Ifl988cX49NNPMXPmTJhMJrz77rv485//jIaGBr+Ltffeew/nnHMOTCYT5syZgyFDhqClpQXbt2/HsmXLfNEToOebeUVFBcrLyyP+0NYDDzyA//znP7jwwgtx9tln47PPPsOKFSuwZs0afPHFFygqKpJr16Ny+eWX46mnnsLQoUMxf/585OXlYf369Vi0aBE++ugjfPDBB1Cr/U/nbW1tmDx5MrRaLX70ox/BZrPhlVdewWWXXQZRFLFgwQLftpIk4fzzz8c777yDkSNHYuHChXA6nXj66afx7bff9qvn+uuvxxtvvIG1a9diwYIFQX+rceONN+L999/H7NmzcdZZZ2H16tV48sknUVNTg48//jjkvl966aV44okn8Mwzz2DOnDn9vv7ss88CgG9/uru7MXnyZOzatQtnnnkmZs+eDUmSsHfvXrz55pv40Y9+hMrKypDvC/RcgJ5xxhloa2vDT37yEzgcDrz66qu47rrr8P333+Oxxx4L63ViJUkSfvzjH+PNN99EVVUVFi5cCIfDgaeeegpbt24N+Byn04nZs2fj/fffx6hRo3DxxRcjKysLq1evxjXXXIMvvvgCzz33XL/n1dbW4qSTTsK4cePw61//GocOHcKLL76ImTNn4l//+hcuvPBCAD3zkpeXhzfffBPnnXcexo8f73uNvLw8v9eM5PxElBQSEcUVACnQUvvggw8kQRAkQRCkPXv2SE6nU6qqqpJ0Op20Zs0av20PHjwolZaWSiUlJZLNZvM9vnjxYgmAlJOTI33zzTd+z7noooskANKLL77oe8ztdksjRoyQAEjvvvuu3/aPP/64r9bVq1f7Hq+trZUASAsWLAi4f1OmTOm3f6tXr5YASIsXLw647YQJE6Tm5mbf411dXVJVVZUkiqJ06NAh3+Pz58+XAEhbtmzp976NjY1+//bWWV5eHrDOQLzjp9FopE2bNvl97frrr5cASJdddlnM+xuu8vLyfvWvWLFCAiDNmzdP6u7uDlj/gw8+6Pe4dx4vv/xyyeVy+R7/9ttvJZVKJY0ZM8Zv+2effVYCIJ166qmS3W73Pd7a2iqNGjVKAiBNmTIl4Hv3PlZ6W7BggQRAGjZsmLR3717f406nUzr11FMlANIXX3wRakgkSZKko446StJqtVJTU5Pf4zabTcrPz5eKi4slp9MpSZIkrVy5UgIgXX/99f1ex263Sx0dHWG9Z3l5uQRAmjx5st+aa25uliorKyUA0tq1a32PD7ROvONQW1sbctuBPP/88xIAadKkSZLVag1Yy0Dzs3DhQr9jwOVySZdddpkEQHrjjTf61QRAuuGGG/xe68svv5TUarWUl5cntbe3+x73HpsrVqwIWHek5yeiZGFMgihBlixZgiVLluDWW2/Fj370I5x99tmQJAnXX389ysvL8c4772DXrl245pprMGXKFL/nlpaW4sYbb0R9fT0++uijfq997bXX4phjjvF77Fe/+hUAYMOGDb7HPv/8c9TU1GDatGmYOXOm3/ZXXHEFjjrqKLl2N6h7770XBQUFvn8bDAZccskl8Hg8+Oqrr/ptr9fr+z3W98N53g8wBRqfUH72s5+hurra77ElS5YgNzcX//rXv4L+KjreHnroIajVajz11FP9xmHRokUoLCwM2JUkOzsb999/P1Qqle+xsWPHYvLkydi+fTs6Ozt9jz/zzDMAgLvuugtardb3eF5eHhYtWhRT/bfffrtfzlqtVvt+vd772AxmwYIFcDgceOGFF/wef+utt9Da2opLLrmk353xQMeMVqtFTk5ORPXfc889fncuCwoKfGMyUOREbt73ufvuu5GVlRWwlt48Hg8effRRlJSU4IEHHvA7BlQqFe677z4IghDwuMnNzcXtt9/u99gJJ5yASy65BG1tbXj99dcjrj/c8xNRsjAmQZQg3oylIAjIy8vDqaeeissvvxw//elPAQD//e9/AQB79+4N2J93586dAIDt27f3i0qccMIJ/bYfNmwYAKC1tdX32ObNmwEAp5xySr/tRVHEySefjB9++CHSXYtYuPVecskleO211/B///d/uPDCCzFt2jRMnjwZQ4cO7ff8WD7A1PeHD6DnomD8+PFYu3Yttm/f7vdr4ETp7u7G119/DbPZjAcffDDgNjqdDtu3b+/3+MiRI2Eymfo97h3ntrY234Xh5s2bffPfV6BjJRLhznUwP//5z7Fo0SI888wzuPrqq32Pey/ie0c+pkyZgiFDhuBPf/oTNm3ahFmzZmHy5MkYP36830VhONRqdcAx8eanvesp3jZt2gRRFAPORaD+wj/88AOam5sxcuRI3HXXXQFfU6/XBzxuJkyYEPAHhqlTp+KZZ57B5s2b/cY7HHIcA0TxxIthogSR+nzgpq/m5mYAwMsvvxx0u66urn6P9c3oAfDdKXO73b7HvB9aGjRoUMDXHuhxuYVb7/z58/H222/jvvvuw1NPPYW///3vAIDjjz8e99xzD84880xZ6hlov0tKSgAg7A97ya21tRWSJKGxsTHiDwwGGmNg4OOioKCg391VIPZjIty5Dmbo0KE4/fTT8cEHH2D79u0YM2YMGhoa8N5772H8+PE47rjjfNuaTCasX78eixcvxsqVK/H+++8D6PlNwlVXXYXbbrsNGo0mrPc1m80BL6ATfVx45ydQ3d5aevOeS3bu3Bn0uAl0LonHWpDjGCCKJ8YkiBTC25/0zTffhCRJA/7X+0NjkfLeKTx8+HDArwd6XBR7ThMulyvgc9ra2qKuJxznnHMOPv74Y7S2tuKjjz7Cb3/7W3z77bc499xz8d1338nyHgONh7ebRLJ6x3rft7q6OugxEeoHrVBMJhNaWloCzvFAY5No3ruR3rvBzz//PFwuV8C7lEOHDsU//vEPNDQ0YNu2bXj44YdRWFiIO+64A3fccUfY79nU1BTwYi3Rx0Vubi5aWlrgdDoHrKXv9gAwb968oMdMbW1tv+cqdS0QxRMvhokUYtKkSQB6uj3EizcX+9lnn/X7msfjweeff97v8fz8fADA/v37+32to6MjIbEKoCdXPH36dNx///245ZZb4HA4sGrVKllee+3atf0ea29vx5YtW3xtzJLBaDTi6KOPxrfffouWlpa4vU91dfWA8x/oWAHgu2OaqDt78+fPh8lkwj//+U94PB4888wzUKvVuPjiiwd8jiAIOProo3HNNdfggw8+AAC88cYbYb+ny+UKOCbe1oN9c+bxMmHCBHg8noBzEegvAI4ePdrXcSTQBXQwmzZt8suT932f3vuc6GOAKF54MUykEOeddx6qqqrw2GOP4d133w24zX//+9+Y/vrZ5MmTUVVVhdWrV/e7kHziiScCXtjm5ORg9OjRWLdund+dWLfbjd/97newWq1R1xPKRx99FPD1vXevsrOzfY85nU7s2LEDu3btivh9nnvuuX75zyVLlqC9vR0XXXRRUls//e53v4PD4cBll10W8C58a2srNm3aFNN7/PznPwcA3HbbbX69ntvb23HnnXcGfE5hYSEAYN++fTG9d7j0ej1+/OMf4+DBg3jggQfw9ddfY9asWSguLvbbbtu2bQFb6wU6ZsJx8803+32AsqWlxZfD9X4QMN6873PrrbfCZrMFrKU3tVqNa665BocOHcK1114bcA0dOnQo4G9W2tvb+909/+qrr/D8888jNzcX8+bN8z2e6GOAKF6YGSZSCI1Gg9deew0zZszAOeecg5NPPhnjx49HdnY29u/fjy+//BK7d+/GoUOHIv6G7iWKIpYvX46zzz4bc+bM8fUZ/uabb/DBBx9g5syZWLVqlS8a4fWHP/wBl19+OSZPnowLLrjA16/U6XTiuOOOw9dffy3HEPTz+9//Hnv27MHUqVMxfPhwaLVabNy4ER9//DHKy8vxk5/8xLftwYMHMWbMmKj6DM+cOROTJ0/Gj3/8YwwePBifffYZPvvsMwwfPhx/+tOfZN6ryFx22WXYuHEjli1bhqqqKsyYMQNlZWVoaWlBbW0tPvnkE/ziF7/A3/72t6jf4+c//zleeOEFvPfeexg3bhzmzJkDp9OJV199FSeccAK+//77fsfEtGnTIIoibr75Zmzbts33G4Tbbrstpv0NZsGCBVi+fDluvvlm37/7+vDDD/G73/0OJ598MkaPHo3i4mIcOHAAb775JkRRxB/+8Iew32/w4MGw2+1+Y/LKK6/g0KFDuOqqq3DaaafJtm/BXHTRRXjxxRexcuVKjBs3Duedd56vlhNPPDHgD4CLFi3C119/jb/97W946623MH36dAwZMgQNDQ3YuXMn1q1bh//3//4fxo4d6/e80047DcuXL8cXX3yByZMn+/oMezwe/P3vf/f7UOZJJ52E7OxsPPjgg2hpafHlja+55hrGKSil8M4wkYIce+yx+Prrr3HTTTehvb0dK1aswOOPP46NGzeiuroazz33XL+WYpGaOnUq1q5di6lTp+Kdd97Bww8/DKvVitWrV/v+GEHfLgSXXXYZli9fjtLSUjzzzDN46aWXcPLJJ2PdunUDflBLDrfccgtmzpyJb7/9FsuXL8ff/vY3HD58GLfccgu+/PJL3wVYrH77299i2bJl2LJlCx588EHs2LEDl156KT7//PN+dx6T4bHHHsNbb72Fk046CR9++CHuv/9+rFy5Eu3t7fjDH/6A66+/PqbXFwQBr7/+OhYtWgSn04lHHnkEb775JhYsWOD7wxJ9j4kxY8bgmWeeQUlJCZYtW4ZFixbF3IYtlFNOOQUjRoyA0+lEQUEBzj333H7bzJgxA9dffz1sNhvefPNN3Hffffjkk09w5pln4tNPP8WPfvSjsN9Pq9Xiww8/xFlnnYUXXngBf//735Gbm4uHHnoIjz76qJy7FpQgCHj55ZexdOlSX9u0lStX4he/+AVeeumlgM/RaDR444038Oyzz2LUqFG+D6K+99578Hg8uPPOO3HJJZf0e15FRQU+//xz5Ofn429/+xteeuklTJgwAe+++67vD2545efn49VXX8XYsWOxYsUK3zHADhGUagQp1k9eEFHamDx5Mr744gu0t7fDYDAkuxxSgA8++ABnnXUW/vjHP+Kee+5JdjkUJ96/4LhgwQI8/fTTyS6HKKF4Z5gow3R3dwfMnj799NP4/PPPcdZZZ/FCOAPV1dX1e6y5uRl//OMfAcAvK0pElE6YGSbKMPv27UN1dTXOPPNMjBgxAi6XC5s3b8Znn32GvLw83HfffckukZLgd7/7Hb7++mucfPLJKCoqwoEDB7Bq1Sq0tLTg17/+NSZOnJjsEomI4oIXw0QZZtCgQbjkkkuwdu1arF69Gna7HSUlJfjFL36BW2+9FVVVVckukZJg/vz5OHz4MN566y20tbUhKysLRx99NC677DL88pe/THZ5RERxw8wwEREREWUsZoaJiIiIKGPxYpiIiIiIMhYvhomIiIgoY/FimIiIiIgyFrtJHNHW1oa1a9di2LBh0Ol0yS6HiIiIiKJgt9uxf/9+TJkyJay/ksqL4SPWrl2LuXPnJrsMIiIiIpLBG2+8gfPOOy/kdrwYPmLYsGEAegZuxIgRSa4mPXV1dWHDhg2YOHEijEZjssuhPjg/ysb5UTbOj7JxfpRN7vmpqanB3Llzfdd2ofBi+AhvNGLEiBE4+uijk1xNeuro6EB9fT3GjBkDk8mU7HKoD86PsnF+lI3zo2ycH2WL1/yEG3vlB+iIiIiIKGPxYpiIiIiIMhYvhomIiIgoY/FimIiIiIgyFi+GiYiIiChj8WKYiIiIiDIWL4aJiIiIKGPxYpiIiIiIMhYvhomIiIgoY/FimIiIiIgyFi+GiYgyUJfdhf0t3eiyu5JdChFRUqmTXQARESWO0+3Byi11WFfThE67Czk6NSaPMGPO+FJoVLw/QkSZhxfDREQZZOWWOry++SC67C7k6jXY3WlBQ6cdAHD+8UOTXB0RUeLxNgARUYbosruwrqYJXXYXRpfkoDRPj9ElOX6PExFlGl4MExFliFaLA51H7ggLggAAEAQBuXoNOu0utFocSa6QiCjxeDFMRJQh8g1a5OjUaLc6IUkSAECSJLRbncjRqZFv0Ca5QiKixGNmmIgoQxiPfFiuodOOHfWdyNVr0G51+h436vgtgYgyD898REQZZM74UgDwdZOoNBt83SSIiDIRL4aJiDKIRiXi/OOHYsa4ErRaHMg3aHlHmIgyGs+AREQZyKhT8yKYiAj8AB0RERERZTBeDBMRERFRxuLFMJHCdNld2N/SzT+AEAGOGRERRYuBMSKFcLo9WLmlzvcp/5wj7a7mjC+FRsWfWwPhmBERUax4MUykECu31OH1zQfRdeQvhO3utKCh0w4AOP/4oUmuTpk4ZkREFCveOiFSgC67C+tqmtBld2F0SQ5K8/QYXZLj9zj545gREZEceDFMsmBmMzatFgc6j9zdFAQBACAIAnL1GnTaXWi1OII+P9XGX456Yx0zIiIigDEJihEzm/LIN2iRo1Njd6cFg3OzIAgCJElCu9WJSrMB+QZtwOel2vjLWW+0Y0ZERNQbL4YpJsxsysN45KKwodOOHfWdyNVr0G51+h4f6I8jpNr4y1lvtGNGRETUm/JuHVHKYGZTXnPGl2Je9RBUmg3QqkVUmg2YVz0Ec8aXBtw+1cY/HvVGOmZERER98dYJRS2czCbvzoVPoxJx/vFDMWNcCVotDuQbtEHHL9XGPx71RjpmREREffHOMEXNm9lstzohSRIA+DKbOTo1M5tRMurUGFaQHfKiLtXGP571hjtmREREffE7B0WNmc3kSrXxT7V6iYgoM/C7D8XEm830dgeoNBt83QEo/lJt/FOtXiIiSn+8GKaYMLOZXKk2/qlWLxERpT9+FyJZGHVqXtQkUaqNf6rVS0RE6YsfoCMiIiKijMWLYSIiIiLKWLwYJiIiIsXqsruwv6VbcX9IiNIHQ3tERESkOE63Byu31Pm6z+QcacM4Z3wpNCreyyP58GKYiIiIFGflljq8vvkguo785crdnRY0dNoBAOcfPzTJ1VE64Y9WREREpChddhfW1TShy+7C6JIclObpMbokx+9xIrnwYpgoA2Ri5i4T95komFRaE60WBzqP3BEWBAEAIAgCcvUadNpdaLU4klwhpRPGJIjSWCZm7jJxn4mCScU1kW/QIkenxu5OCwbnZkEQBEiShHarE5VmA/IN2mSXSGmEF8NEaSwTM3eZuM9EwaTimjAeuWBv6LRjR30ncvUatFudvsf5R3tITsr8kZCIYpaJmbtM3GeiYFJ5TcwZX4p51UNQaTZAqxZRaTZgXvUQzBlfmuzSKM3wRyuiNNJ1JEuXb9CGlblLt7srmbjPRMGk8prQqEScf/xQzBhX4juvKbVWSm08qojSQKBM4PHl+TBoVWjstGdM5o45QyJ/6bAmjDo1L4Iprnh0EaWBgTKBw/L1MOrUGZO5Y86QyB/XBFFoXAVEKa5vJlAQBAzOzcKO+k5YnW7MPKYEG/e0otPuQqXZ4PsUebry7pv3Lnkm7DNRMFwTRMHxYpgoDL2zuEq7kxIsE2hxuHHayCKcN36IYuuXG3OGRP64JoiC42ogCiIV+nOGkwnMxMxdJu4zUTBcE0SBcVUQBZEK/TmZCSQiIooev0sSDSBYFnddTRNmjCtRzIUmM4FERETRUcZ3ciIFSqX+nMwEEhERRUcZoUciBfJmcdutTkiSBAC+LG6OTq3I/pxGnRrDCrJ5IUxERBQmfsckGgCzuEREROmP382JgmAWl4iIKL3xYpgoCGZx40PJfZvTBcc4c3HuiSLDVUIUBvbnlEcq9G1OdRzjzMW5J4oOv7sTUcKkQt/mVMcxzlyce6Lo8EdFIkqIvn2bS/P0GF2S4/c4xYZjnLk490TR48UwESVEOH2bQ+myu7C/pZvf2AcgxxhTauLcE0WPMQkiSghv3+bdnRYMzs2CIAi+vs2VZkPQvs3MQoYnljGm1Ma5J4oeL4aJKCFi6dvMLGR42Bs7c3HuiaLH1UFECRNN3+a+WUhBEDA4Nws76juxrqYJM8aV8Bt9L+yNnbk490TR4XcQIgooHr1Ko+nbHE4WkhfD/8Pe2JmLc08UHa4SIvLj8kh4deOBuOZzI+nbzCxkdNgbO3Nx7okio6hPnnz88ce47LLLMHr0aBgMBgwZMgTnnXceNm7c2G/bTZs24YwzzoDRaEReXh7mz5+P3bt3J6FqovSyekcDXt98ELubLHC4PNjdZMHrmw9i5Za6pNTTO/O4o74TdW1W7KjvZBaSiIhkoaiL4ccffxx79uzBddddh3fffRcPPfQQGhoaMGnSJHz88ce+7Xbs2IGpU6fC4XDgpZdewlNPPYUffvgBp556KhobG5O4B0Spb/O+VsX1Kp0zvhTzqoeg0myAVi2i0mzAvOohzEISEVHMFHVL5bHHHkNxcbHfY2effTZGjBiBu+++G9OnTwcA3H777dDpdHj77bdhMpkAAMcffzxGjhyJv/71r7j33nsTXjuRV6isbTRZ3HjkdwdicbhlyefKWTOzkJSqErl2EyHd9ocIUNjFcN8LYQAwGo0YO3Ys9u/fDwBwuVx4++238fOf/9x3IQwA5eXlmDZtGl5//XVeDFNShOqFG02v3GT01zVoVahtd0adz41nzcxCUqpIt97Y6bY/RL0p/rtKe3s7Nm3a5LsrvGvXLlitVhx77LH9tj322GPxwQcfwGazISsrK9GlUoYL1Qs3ml65yeivW12Wjz0dLVH3KmVPYKL0Wwfptj9EvSn+Yvjqq6+GxWLBrbfeCgBobm4GABQUFPTbtqCgAJIkobW1FYMHDx7wNRsaGvpli2tqagAAXV1d6OjokKt86sVisfj9bzrpdrixeVcd1G4bqksMPXdUc0XUNlmweVcdJpRmBf365HIDsrWqiF4z0HNi4Z2XicMMAHqywxaHG8OMGlSX5WNaVU7ItZHomjNJOq+fdNB7ftJtHaTD/nD9KJvc89PV1RXR9oIkSZIs7xwHixYtwl133YVHHnkECxcuBAB8/vnnmDx5Ml544QVceOGFftvfc889uOWWW3Do0CGUlJQM+LpLlizB0qVLA37t4YcfRllZmXw7QUREREQJs2/fPlx77bXYtm0bjj766JDbK/bO8NKlS3HXXXfh//2//+e7EAaAwsJCAP+7Q9xbS0sLBEFAXl5e0Ne+6qqrcMEFF/g9VlNTg7lz52LixIkYM2ZM7DtA/VgsFmzYsAETJ06EwWBIdjmy6na48ejHO3Gg1YoKs8GXta1tsmBovh6Xn1qJf3y6e8CvL5w+MuCd4WCvGeg5sZBjfhJdcyZJ5/WTbC6PhNU7Gny/DTFoVT2/DRldDLUohPUavedH0GTFZR3IUWc0YlnXyaq5L64fZZN7frZv3x7R9oq8GF66dCmWLFmCJUuW4JZbbvH7WlVVFfR6PbZu3drveVu3bsWIESNC5oWLi4sDflgP6PnAXu8P5pH8DAZD2o2xCUB1VSn2bD6IzfX2XlnbLFRXlaJisBnVVbYBv15izo/4NQM9Rw6xzE+yas4k6bh+ku3VjQfw+rYWXx62tt2JPR0tgEYfcR7WOz/xWAdy1hmJWNZ1smoeCNePssk1P0ajMaLtFXcxfOedd2LJkiW47bbbsHjx4n5fV6vVmD17Nl577TX8+c9/Rk5ODoCeW+KrV6/Gb3/720SXTAQAvp633k9bV5oNvk9bh/P1aF5TiVKxZspcvXtojy7JgSAIGJybhR31nVhX04QZ40qi6mAi9zqIV53himZ/kl0zUbgUdRTed999uP3223H22WfjnHPOwfr16/2+PmnSJAA9d45PPPFEnHvuufjjH/8Im82G22+/HWazGb///e+TUToR7C4PJlYU4JSRZjhcnn59OKPplZuK/XVTsWbKXK0WBzqP3LWMtbd2b3Kvg3jVGa5Q5zcl1iwn9ldOb4qa0bfeegsA8N577+G9997r93XvZ/1Gjx6NNWvW4KabbsKPfvQjqNVqTJ8+HX/9619RVFSU0JqJgvXfDCSaXrmp2F83FWumzJNv0CJHp8buTkvUvbWDkWsdxLvOgUR6flNCzXJif+XMoKjvVGvWrAl72+OPPx4ffvhh/IohChP7bxKlLm8P7YZOe9S9tRMhWXXGcn5LlbENhuf3zKD8I5FIwZiJI0p9qZJzT3SdcpzfUmVsA+H5PXNwFolikE6ZOJIfc4apIZk590iOkUTXKcf5LZU/Q8Dze+bgLBLFIB0ycSQ/5gxTUyJz7rEcI4mqU87zWyp+hoDn98yRWkcmkcKkQyaO5MecIYWSCsdIpp/fMn3/MwlnkihGqZyJI/kxZ0ihpNIxkunnt0zf/0yhjNVGlMKUkIljNlU5mDNUvmSuly67C9/VtaPV6kyJY0QJ57dkyvT9zxScUSKZJCMTx2yq8jBnqFzJXC+937u124Gaxi4IAAaZdFCJouKPkVTM/Mop0/c/3XFmiVJYKuQOMw1zhsqVzPXS970FCOiwOrF+dzMqzEYeI0RJxFtHRCmqb+6wNE+P0SU5fo9TcswZX4p51UNQaTZAqxZRaTZgXvUQ5gyTKJnrJdB7T6oogEmvgQRAJQo8RoiSiD9+UlrKhAxtJmZT+86rUueZOUPl6b1e3B4JdpcbOrUo23rpdrh9/2sK8t7etapSiagwGyAKwJVTKjG2NDdux4hS1wmRUnBVUFrJpAxtJmVT+86rQauCXqOC1emGxeFW7DwzZ6gc+QYtDFoVvq/vwJ6mLrg8gFoEJAATyvKjXi/eY3PzrjqcqAMe/XgnqqtK/Y7FUGs1XhfCmXQ+JIoFz9KUVjIpQ5tJ2dS+8/p9fSc6rE6Y9GpUmI1pPc8kD6NODb1GhQ6rCw63Bzq1CLvLA61KhF6jinq9eI9NtduGE0uAA61W7Nl8EMD/jsVkrdVMOh8SxYI/GlLayMQMbSZkU/vOa1GODqIAONweCACKc3RpP88Uuy67C1anGya9BsU5OmRr1SjO0cGk18DqdEd13PQ+5irMBgBAhdkQ8FhM9FrNxPMhUbTS59YRpYxuhxvtLd2y59fikaFVetYu3bOpvp6s3Q7fvNpdbjjdEnRqES4PYHd5YNCp0zorHQ2lH7uJ1mpxwOJwo8JsQFGODnZXz93hxk47LA53VMdNJOecvmtVqxbhcHlgd3niElnIxM8UEEWLK4ESxuWRAPRk6g7bRNnza3JmaFMta5du2VS/nqxW55GerAIG5eigU4vQqATYXR6YsgCdWvk9WhMp1Y7dROl7ftDo1DEfN71fU8rtGdtQr6lTi9hQ2xL3+cmkzxQQxSp9vnuS4q3e0QA1ejJ1kkYve35Nzlwes3bJ1b8nK3p6sta2oMJsgEcCtCoREoCGTnvaZqWjwWM3sHjkdnu/Zm2TBeNLgNomC4y6rAFfM1Hzk0mfKSCKFVcDJUSX3YXN+1pxoq4nU2cTdBicm4Ud9Z1YV9OEGeNKZDk5y/F35Ptm7QRBiEutFFig8R9k0mH97mZIkCAKwISyPL9uEtHMczrisRucHOeHgV5z8646AG0Ymq/3dZPoK9HzE4/9JUpHmXtWpITy5vWgQ1zza+FmaIPlKSPN2jGbKa+APVlFERVmI1Si4NeTlWPvjznR4OKRsfe+5uRyA75YdwgLp49EiTk/4LaJnp90/0wBkVy4KighvD0+gZ5MHYTQ2bpYDJShDSdPGW7WjtnM+IikJ2u6ZaVjxZxoeOJx3GQfOb95/zeQZM0P1wlRcFwdlBBGnRrVZfnA4UOobbJA0niSkl8LJ68XbtaO2cz4YNYxehw7ZeP8ECkTVx4lzLTRxfj08HcYmq/HYZuY8PxaJHm9UFk7ZjPji1nH6HHslI3zQ6Q8/G5NCaMWezJyC6ePhFPURZ1fizYnGktP0L7vJUf2j3nXgSUj65gu88GcqLLJMT/pcqwSKQVXESVctlYFkyk74ufFmtGNJq83UNYuluwfs8bhS0TWMV3ngzlRZYtmftL1WCVKNp4pKWXEmtGVM68Xy2sxa6wsnA9KFTxWieKDP0pSSuib0S3N02N0SY7f4+GYM74U86qHoNJsgFbdk1ueVz0kqrxeNK8l136QPDgflCp4rBLFD+8MU0qQqz+nnHlK72udMtKMvc0WlBcaMMiUlZD9iFUiM4fxeC+5XlMp80HkNdCxzWOVKH64ciglyN2fU448ZTT5vWT3gQ1WcyLfK9p8o9yvmez5IPIKdWzzWCWKH14MU0pQYn/OaPJ7yd6PYDWfOdKUsPeKNt8o92smez6IvEId2zxWieKHq4dShpL6c8bSZzhZ+xGq5snlhoS9VzR9mOPV21lJxxVlpnCPbR6rRPHBi+EMkuq9KZXUPzWW/F4k+yHnnIWqucPqjOj1gtXW+73cHgl2lxs6tRhTvjFemcm+86FVi3C4PLC7PGxXleESdc5stTjQ2u2ATi3C5ZGgUQkBj+1Yj9VU/x5AFC9cDRkg3XpTKqF/qhz5vWD7EY85C1WzSa8J63XCqS3foIVBq8L39R3Y09QFlwdQi4AEYEJZflT5xnhnJnVqERtqW9JmnVD0EnnOdLo9+GRnI2oau9De7cThDhtKcrNQlp894LEd6bGabt8DiOTGi+EMwN6U8ot3fi8ecxaq5mytSrbajDo19BoVOqwuONwe6NQi7C4PtCoReo0qqvFJxTGn1JTIY2Hlljqs2loPAYBKENDQaUeb1Ym6NhvKCrIDHtuR1sdjmyg4/kiY5tibMn7k7FncWzznLNaaw62ty+6C1emGSa9BcY4O2Vo1inN0MOk1sDrdUe9DKo45pZZEHgu9X3NSZSFGDc5BkVELj0eCBAkzx5X0O7YjrY/HNlFovDOc5lKhN2Wq5tjsLg8mVhTglJFmOFwe2eqP55wFq9l6ZJtuhxvtLd0DZoFbrU7o1CLcHgnqXtnGNqsT39W1o7zQgL3NFrTbXKgwG1CUo4Pd1XN3uLHTDovDLVu+NxXGXCn6rrNUXXfxlshjofd7qUQRFWYjhuVnY19LNww6NU47qqhfjCHS+jLh2CaKFVdAmlNyb8pUzbHFu1dvPOYsnJpdHgkA8OjHO3HYJvabD6fbg09+aERNQyfau52ob7dicJ4eQ/P1qG3qgiQBt7/5LSwOF4xaNTrtToiCgEEmHYw6tazHndy5cSWvk1j1nXuDVgW9RgWr0w2Lw50y6y5REnksBHovlSjA7vJgSJ4m4HtFWl86H9tEcuHFcJpTcm/KVM2xxbvueMxZODWv3tEANYADrVZIGn2/bVZuqcOqbfUQIEAUBTR2OdBhdWHn4S64PRJEEeiyuWB3e9CpdkEUBHg8EtbvbkaF2aiY4y4QJa+TWPWd++/rO9FhdcKkV6PCbEyZdZcoiTwWonmvSJ+Tzsc2kVy4CjKAEntTxqtnbLwlqm455yycmgFg875WnKgDKswG2ASd3zanjDT/L9tYUYB9rd2ob7ehxeKA3elBfrYGKgGwODwoMejQbnXBqFPDLUmQAKhEQRHHXTBKXCex6jv3Lo+Evc0WONweCACKc3Qpse4SLZHHQjTvFelz0vHYJpITz3oZQEn9eb2SnWOLNi+ZqLoHmrMuuwv17baI6g6W8/XWDAAWhxvQIeB+7W22/C/bqOrJNg4yZeGH+k4c7rShKEeHxs6evqeCIEKj6umXOjRfj2ytCldOqcTY0tykH3fBKHGdxKrv8Wp3ueF0S0f62fZkyA06NfOjfchxLIR7jolmrUdan5znE6J0FNPR/+2332LdunU4ePAgrFYrzGYzxo4di9NOOw0mk7x/2pVip4T+vF7JyrHFmlNOdN3eOXO6PXh144GI6x4o51tW0L+HqeFIazVJkgABfvtVXmjw7fegHJ3vznBzlx1uCWjstEMtCuiye6DXeOB0e5ClUaPb4caQPL3iL4R7U9I6iVXf41WnFqFR9WRSTVk9/WqZHx1YNMeCyyNFtVajWeuR1hfr+YQoXUV8xm9tbcXf//53PPHEE9i7d2/PN86+L6pWY9asWbj22msxffp0WQql9JKsHFused9Uq3ugnG9dmxVlBQa/mqvL8oHDh1DbZIGk8fjt1yBTlm+/19e2oMPmhMPlgU4lQiMCFrsbKrEnI1zfYYdWLcIjAbl6DXOJSRToePVIgFYlQgLQ0GlnflRmq3c04PVtLVGfYxLxWYpU/bwGUbxE9CPgww8/jBEjRuCvf/0rZs6ciRdeeAE7d+5Ee3s77HY7Dh06hHXr1uFPf/oTWltbceaZZ+Lss89GTU1NvOqnFBavnrEDkavfZqrU7dfDtKIAo0tyUJyj8+V4Zx7j38N02uhiAMDQfH3A/ZozvhQzx5VAggSPR0KRUYtRg3MwfVQxTHoNsrUqlORmodCoxZA8PSaU5cV1XCg8fY/XCWV5mDmuBBPK8hNy/Gaazftaoz7HJKInMPsOE/UX0W2Ahx9+GA888AAuuugiaDT9/3TroEGDMGjQIJx00kn43e9+h127duH//b//h5deegm33HKLbEVTekh0RlOuvG+i83eB6nZ5enKfTV12fFfXHjCG4NfD9EjOd2h+Nva3dCNbq8JpI/17mKrFntdeOH0knKKu335oVCJOO6oIH+5oQLfdhbKCbKiPPL/CbIAoAL+ZWoXyQoOsfZcpPANlVIMdr+mSjVYSi8ONXL0GLo8Eu8sNnVoM+xwT7TkqkrlM9uc1iJQooiN+x44dUKvDf0pVVRWeeuopuN3uiAujzJGojKbced9E5e961+3N6x5qs6Gh0wZBELD4zW8xqiQHp44s8nvPQPur9vUw1Q+4v9laFUym7AFryddr0GpxQHXk4rn3GKZSNjhdhJuD77vO0ikbrSR6jYjaOgv2NlvgdEvQqAR4JGBCWV7Ic0yk56hoPgPBvsNE/UV0JozkQrg3lUoV1fOI5BSvvG8i+w5787rddhc8EiAKwKEOGywON5q6HH7vGY/9Zc9S5WH+U1my1Cp0WJ1wuHv+6qLd5YFWJUKvUYVcH5Gur2jmnmuYqD9Zjvru7m7YbLZ+jxcUFMjx8kSykbvfZiL7Dtucbvztk13wHGmPJgoiioxadNhcEAWg3ers957x6C/KnqXKkar9utOZzeWGSa+GAMDlAUxZgATA6nSjy+4KOR/hrq9Y5p5rmMhf1GfJ7u5u3HLLLXj++efR0tIScBvGI0hp5M77BsvfNXbZsOb7wzhxeCEGmbJiymj2zuu2Wuxo6LBDAiCKPT197S4P8gSgtdvhl/mLRy7b+5qnjDRjb7MF5YUGDDJlDbh9sP1mbjU28c5/cn78hTMeVqcHFWYjinN0sLt67g43dNphcbjDmo9w12wsc5+OPbWJYhH10b9w4UI899xzmD17NsaMGQOtljkjSh1y5X0D5e+cLje2HmyHzenG9kOd0GtUGJKnR7FJB6vTE3Wm2JvXbe60Q6MS0GV3I0vtRpfNCQk936jzuh34ZGcjfnzCsKB50ViEm1MMth2AmPo9U4945T9j7cedbiIZD4NWhdp2JwbnZsGgU0c9H6HWrBxzz9w4UY+oV8Fbb72Fe+65BzfccIOc9RAlVDz6Dm892I4OqxOiALg9Ig532HGo3QaTXo1jhuRFnens/V5tVgc8HgkH22xHssMCDCoRAoBVW+uRpVbFLS8a7pgF2w4Ac64ySNUcfKqJZDyqy/Kxp6Ml7nlcZn+J5BPTj/jV1dVy1UGUcPHoO+yRPLA53RAFYEieHnnZ2iOfJpdgd3qQn62Jqaen970mlOWjJDcLalXPXxQbnKvDqME5mFRZGNd+oeGOWbDt1vzQgDXfN7DPqUzk7nvNPrT+Ih2PaaOLE9aHPNE9z4nSVdQ/Os6fPx//+c9/cPrpp8tZD1HYYs0ztlocaO12QKcW4fL0tECKJG/Z+/29+bs13x/G9kOdcHtEiKIIp9vTk+0VBLg9ErrsLui16qgznb2zft/VtePRj2tgd3n8ev7K3S+0936Gm1MMtl2bpSfW0bsXqyj0/GngvpnnSGqT6zmplpOVO//JPrT+Ih0PtShENR/RHHfM/hLJI+pVc9999+H888/H7373O8yaNStg54gJEybEVBxRIHLkGZ1uDz7Z2Yiaxi60dztxuMOGktwslOVnh8zcBXv/E4cXQq9VobnLAUnyQCUKEAB4JAkqUYQxhgxhb0adGmNLc2E26rC7yRKw52+s/UID7efxw/Nh0KrQ2GkPmlMMlmccVqCHxyNh68EO7GnqQofNBYfLA48kwWzUBcw8h1NbqGMg1HNSPScrV/6TfWj9RTse4c6HHMcds79EsYl69VitVrhcLjz44IN46KGH/L4mSRIEQWA3CYoLOfKMK7fUYdXWeggAVIJwJIfrRF2bDWUF2UEzd6He/5QqM1Ztq0d9hx06tQinW4IoCNBpRLR2O9Fu7ZYl1xfvzOBA+zksXw+jTh30PYPVNvWoYny1pwUdVie6HT39kj1Szxh1O9xhZZ6jOQZCPYc52R7MovpL1joDMuu4I0qmqFfx5Zdfji+//BLXX389u0lQwsjRV7X3a0yqLMS+lm4carOitdsJCRJmjisZMHMXzvsvmj0WAPDZribYHG4MMun8uknI2dMzXv1Cg+2n1enGzHEl2Li3Neh7DlTb9DHF+HRnI4xZKjjdHjjcHujVIgRBgFHX8wcLgs1lNMdAqOecMtLMfr29sA+tv2Sss0w87oiSJepVtnr1atx///341a9+JWc9lCKSlauMNc/YZXfhu7p2tHY7kKvXQCWKqDAbMSw/G/taumHQqXHaUUUD/noynPcfVpCNe84/Foc7bH59eOMxZpFmBsOtIdh+WhxunHZUEc6rHhK0V/NAte1v6YbF4caQvGx4PN1wuSUYs9SwOt1weQC9VhV0LqM5BkI9Z2+zhTnZXphF9SfneESTwSei+Ip6leXk5GD48OEylkKpINm5ymjze73rbrU6UdPYBQECBuXooFKJUIkC7C4PhuRpgmYiI3n/QaYsvz9GEc9cX6jXjnTewtnPcHs1963N+9r17TZo1SLsLhcACQ6XB0adClaHG0Py9APOQzTHQKjnlBcamJMNgFlUf7GMR8AMfnl4GXwiiq+oz3I///nP8cILL+DMM8+Usx5SuGTn26LN7/WtWwDQYXVifW0LKsyGsDOAqZqnjHTeItnPWF673eqEW5J68tUqERIAk14TdCyjmYNQzxlkykrJeaXUEUsGn4jiK+qVdtxxx+HWW2/FvHnzcM455wTsJjF//vyYiiNlUUq+LdL8XqC6B5l0WL+7GRIkiAIiygCmWp4y2nkLZz9jfe1PdzZiZ0MXOmxO5GZpMKLYiFNHFoUcy2jmINRzUm1eKXWEzOAfU4KNe4Jn8IkofqK+crnkkksAAHv27MGbb77Z7+vsJpF+lJJvizS/F6hub1ZYJQq4ckolxpbmytbbU2l9aqOdt3DGWa7Xdro9aOqy+/LVoUST4bS7PJhYUYBTRprhcHn6PYc52eRQ2nqJh5AZ/JFFOG/8kLQfByKliukDdJRZlNZ/NNz8Xqi6I7kQDvb+yc5TDyTWeQs2zrG+tk4tYkNtS9RjFs4xEGxeon1Nip1S10s8RJLBJ6LEi3rlTZkyRc46KAWkal42UXUnO089kHjuf6yvnYgxU+q8ZLpMmpdUPXcSZQquQIpIquYq4123UvLUA4nn/kf72okYM6XPS6bKxHlJ1XMnUSaI6Wzzxhtv4Pnnn8fevXths9n8viYIAr7++uuYiiPlSdVcZbzrVkqeeiDx3P9oM9SJGDOlz4sSJCOzq7R5CTUGcoxRvHoVZ/rxSySHqFfRX/7yF9x0000oKirCiBEjYDAY5KyLFC5V823xqltpeeqBJLLXcahMaCLGLFXmJRmSmdlVyryEGoN4jJHcvYrTNWdNlEhRf1dctmwZLrvsMvz973+HSqWSsyailMNMYH+hMqGJGDPOy8CSmdlVyryEGgOl5ZqVVg9Ruoj6jNPc3IyLL76YF8JERzAT+D/hZkITMWacl/6UkNlN9ryEGoNTRpqTPkaR1JuOOWuiRIl65UyePBnbt2/H9OnT5ayHFCYRWToli2T/vJnAU0aasbfZMmDP3GSMWaLfs3cm1OWRYHe5oVOL/TKhicigy/Ue6XSsKyGzm+zPH4Qag73NlqSPUST1hlNPOh3DRHKKejU8+OCDmDdvHoYNG4azzz4bWm3mZu/SUTKydEoSzf4pccySNU/5Bi0MWhW+r+/E3mYLnG4JGpUAjwRMKMvrlwlNRAY92vdIx2NdKZldIHmfPwg1BuWFBsWMUTj1BqsnHY9hIjlFfQYaMWIEzjjjDMybNw+CICA7O9vv64IgoL29PeLX7ezsxJ133oktW7Zg8+bNaGpqwuLFi7FkyRK/7SRJwvLly/G3v/0NO3fuhEajwbhx43DjjTfinHPOiXa36IhUy9LJLZr9U+KYJWuejDo19BoVOqxOONwe6NQi7C4PtCoReo0qpe5KpeOxrpTMbjKFGoNBpixFjVEsc5aOxzCRnKJezTfeeCMeffRRjB8/HmPGjJHtznBzczOeeOIJHHfccZg7dy6WL18ecLvFixfjzjvvxJVXXok//elPsNlseOSRR3Duuefi1Vdfxfz582WpJxOlWpZObtFk85Q4ZsnMGHbZXbA63TDp1RAAuDyAKQuQAFidbnTZXSlxjKRzTjPZmV0lCDUGShujaOpJ52OYSC5Rr4Cnn34aN910E+655x4560F5eTlaW1shCAKampoGvBh+6qmncMopp+Dxxx/3PXbmmWeipKQEzzzzDC+GY6C0LJ0cObdIXiOabJ7Sxiza/QgkmvFvtThgcbhRYTaiOEcHu6vn7nBDpx0Wh3vA91ZapjHWMVTa/vSW7MyuEoQaA6WNUTT1KCEfTqR0Ua8At9uNM888U85aAMC3WEPRaDTIzc31eywrK8v3H0VPKVk6OXJu0bxGNNk8pYxZrPvRWyzj3/e9DTp10PdWaqYx2jFU6v4Ekqo9w+UUagyUNkaR1KOkfDiRUkW9us866yysX78+ad0krrvuOtxwww34xz/+gfnz58Nms+Evf/kL2tvbce211yalpnShlCydHDm3aF4jmmyeUsYs1v3oLZbxj/S9lZppjHYMlbo/lHmYDycKLepVsGjRIlx44YUwGAw455xzUFBQ0G+bQI/J5frrr4der8fVV1+NX/7yl773e+uttzB58uSgz21oaEBjY6PfYzU1NQCArq4udHR0xKfoFDKtKgdwFmDzvlZYHG4MM2pQXZaPaVU56OjoCPn1QCwWi9//BtPtcGPzrjqo3TZUlxh67mbkiqhtsmDzrjpMLjcgWxu8x3UsrxHN/sVjzGIVyXv2nh85xj/c95bjveIp0nmL1/5Esn4o8ZQ8P8k49yiNkueH5J+frq6uiLYXJEmSonkjUez5VV+wWIPb7Y7mpX2amppQVFQUsJvEihUr8Jvf/AYLFy7EzJkz4XA48Oyzz2LlypV47bXXMGPGjAFfd8mSJVi6dGnArz388MMoKyuLqW4iIiIiSo59+/bh2muvxbZt23D00UeH3D7qO8O333572PleubW2tvruCP/1r3/1PT5z5kxMnToVV155JWprawd8/lVXXYULLrjA77GamhrMnTsXEydOxJgxY+JWeyayOd3425pd2F3fjHlDbXj9QBYqSwpx5dQqZGkC3yHrdrjx6Mc7caDVigqzwZdzq22yoDQvC8cMycW3dR2wONwwaFU9dzlGF0MtCmG9xtB8PRZOH5nUO45eLo+E1TsafHdtBtqfeL+HxWLBhg0bMHHiRAiarH5j5/Z48M2BdkiQUFZggClLLUudqTJP4YrX/vSeH4PBEIfK++t73Og1IrLUKthcblidnrgcq6mqvbMLm776ElsdZjTbxYjHJhHngUyWjPVD4ZN7frZv3x7R9lFfDPe9U5tI33//PaxWK0488cR+XzvhhBOwdu1adHV1wWg0Bnx+cXExiouLA37NaDTCZDLJWm+mu+fVb7BqWwvyNG5gKLCn3Y0tTS2wQot7zj824HNMAKqrSrFn80Fsrrf3yrllQVLr8c6Odl8es7bdiT0dLYBG75fHDPYa1VWlKDHnJ2YAQnh14wG8vq0l5P4k6j0MBgNMJlO/sattsqDD6oFJr0auS419jfLUmSrzFK547493fhKh33FTZ0GH1QmTXo0KszEux2qq+uC7w1ADqGlxQdLoIx6bRJwHKLHrhyIn1/wMdP03EGV9rDlMpaU9PRXXr1/v97gkSVi/fj3y8/P5k59CHO6w4bNdTXC4PTAbez61bDZq4XB78NmuJhzusA343DnjSzGveggqzQZo1SIqzQbMPKbE16d2dEkOSvP0GF2S49dLM9RrzKseopheqn17gIban0S+R++xEwVAggSTXo1JlYWy16n0eYpUOuxP3+OmKEcHUQAcbg8EAMU5OtmP1VTVZXdh875WAECF2RDx+kjEeYCIBhbRneE///nPuOaaa6DX68N+zsaNG1FfXx/RX4VbtWoVLBYLOjs7AQDfffcdXnnlFQDArFmzUFZWhvnz5+OJJ56ATqfDrFmzYLfb8cwzz2DdunW48847kxbhIH97my2wOdzQqUUIgjdnLkKnFmFzuLG32YJBpsCt8AL11Gy1OLDm+0bk6jVweSTYXT2vPVDPzL6voVWLcLg8sLs8EbW4ile/2Eh7gEZaR5fdhe/q2tFqdUbcZ7T32H1X147H1+6GxyNB1evzAnL1KlVaP9dYJWJ/4t3DuNXiQGu3Azq1eGSteeBweaBRCXC6e/5t0KnZrxb/66sNHSJaY72fz17AFA9K7nWuJBGNzIoVK/DAAw/g8ssvx09/+lOMHj064HY2mw3vvPMOnnrqKXz44YdYsWJFREX95je/wd69e33/fvnll/Hyyy8DAGprazF8+HA8//zzePTRR/Hcc8/hqaeegkajwVFHHYV//vOfuPjiiyN6P4qf8kIDsrQqNHc5IEkeAIAk9VyMGo1alBeGvoPft6emQavC9/Wd2NtsgdMtQaMS4JGACWV5A/bM1KlFbKhtibjva7z7xYbbAzTSOnpv39rtQE1jFwQAg0w6qEQxoj6jRp0aY0tzka/XYHdTfHuVKq2fa6zisT8uj4RXNx6Iaw9jp9uDT3Y2oqaxC+3dTtS32yCKQFu3AzaXB063B4faezLR7Ffbs44NR3LgkiQBAiJaH+wFTHJLpV7nShDRWXrr1q147LHH8Ne//hX33HMPiouLMWHCBBQXFyMrKwstLS3YtWsXtm7dCpfLhXPOOQebNm0K65N8ve3ZsyfkNllZWbjhhhtwww03RPTalFiDTFk4pcqMVdvq0dTlAAA0dTmgValxSpV5wLvCAzHq1NBrVOiwOuFw9/xVM7vLA61KhF6jkr3va7z7xYbbAzTSOvpuL0BAh9WJ9bubUWE2RtxnlL1KlWP1jga/bGk8ehiv3FKHVVvrIQBQCQIOtdvgkSRIkgRREOByS6g53IXDHTaUFRgy/hgw6no+TIrDh1DbZIGk8US0Pri+SG7sdR6ZiFaYWq3Gddddh4ULF+LNN9/Eu+++i//+97/4/PPPYbVaYTabMXr0aCxatAgXX3wxKisr41U3pZBFs8cCALbvPwzAibxsDU4aNsj3eCS67C5YnW6Y9GoIAFwewJQFSIAvS9z3G0ffPJ4gCBicm4Ud9Z1YV9OEGeNKBvzTwNE8L1LeHKn3J/hKs8H3E3w0dQTaflCODutrWyBBgkoU+r2HHHVSYmze1xrXY7L38TOpshC1TRZ8W9cOm1NClkbEoJwseCSgtdsBCcDMY0p4DACYNroYnx7+DkPz9ThsEyNeH1xfJJdEfe9KJ1GNhkqlwvz58zF//ny566E0lK1V457zj0XtoSZ8s2EdHriwGhWDzVG9ljebV2E2ojhHB7ur5+5wQ6cdFoc7YLYu2jxeonJ8ofKlkdYRaHsPgMG5WRBFYP6EUpw4vDDiu/LplutNVRaHO67HZO/jRyWKGJyrx8E2Kyx2FwxaNcYNzYNOLWJ/SzeytSqcNrKIv3YFfO3PFk4fCaeoi3h9cH2RXJhBjxzPYJQwhUdyb4Ux5N+82bp2qxMqUYBBp4ZKFNBudSJHpw6Yrev9HO/fmPHm8QZ6TizPi5ZRp8awgux+J6lI6+i9vdvtQW1TF76sbcbXB9rw9f52PPbxLtzz7na8uvEAnG6PbHVSYhi0qrgek32PN51ahEbs+dCcVt3z4Ve1KMDu8iA/W8s8ax/ZWlVM64Pri2KV6O9d6YCrjVJKNNm6aPN4SsnxRVpH7+3X17agw+ZEt90FjwSIgoT6dhssDpcvw838WGqpLsvHno6WuB2TgY43CYBWJcIjAY2dduZZiRRMKd+7UglHhFJONNm6aPN4SsnxRVrHnPGlPX/575NdcLt72sgJAlBk1KHD5oIAoMPqZH4sBU0bXQxo9HE9JvsebxPK8qHXqGB1umFxuJlnJVI4pXzvShX8DkgpJ5psXbR5PDlzfLH0e7S7PJhYUYBTRprhcHlCvoZGJeK0o4rw4Y4GtHbZ0dBp77kzLIrQqkW4PIBeq0pIfkypfS6VWlcoalGIe7Z0oOM+VceMKNMwgx4ZjgylrGh6uEbb9zWWfrGx9HsM9txQ8g1a5Os1aOq0Q6sW0WlzQZJ6/nCCUaeC1eHGkDx93PJjSu1zqdS6IpWInsx93yPd+kATpTuu2fBwhIjiLJZ+j7E8t3durN3qhFuSUN9hh04lQgJg0mvimh9Tap9LpdZFRETJkTq3QYhSUN9+j6V5eowuyfF7PB7P9ZozvhTzqodgQlkehuTpUWjUYki+HhPK8jGvekjc8mNy1J5JdRERUfJEfUvI6XTi3nvvxb/+9S/s3bsXNpvN7+uCIMDl4jcWipw3l6hVi2HlY5Usln6PwZ7bZnXiu7p2jC3NDZkd7p0bS9SY9q7d5ZFgd7mhU4tJ73PJ/pvBpXImOJVrJ6LkivqMcfPNN+OBBx7AzJkzMXfuXOh0OjnrogzkzXJ+urMROxu60GFzwpSlwchiI04dWZRymU7gf/0ed3daMDg3C4Ig+Po9VpoNQfO6gZ7r9vT0DZYAPL5mF/KztWHlXROdG8s3aGHQqvB9fSf2NlvgdEvQqAR4JGBCWV7S+lzGMh/pLJVz1KlcOxEpQ9TfHV966SXcfvvtWLx4sZz1UAbzZjn3tXSjw+aEw+VBl9WFTpszZXvixtLvMdBza5u60GF1waTXwCMBu5uUmXc16tTQa1TosDrhcPf8lUC7ywOtSoReo0ranTv23wwslXPUqVw7ESlD1D82t7a24rTTTpOzFspg3sxmu9UJUQBUgoASkw6iKPj1xE3FTKc3t1tpNkCrFlFpNoSd1+39XJUo+D74NqmiQNF51y67C1anGya9GkVGLbK1Pf9r0qthdbqTWmss85GOUjlHncq1E5FyRH0b5LTTTsOWLVswbdo0OeuhDOXNcmZrVWixOI78kYjE98SNh1j6PfZ+7nd17Xh8zS54JEB15Ne/ofKuycpRtlocsDjcqDAbUZyjg93Vc3e4odMOi8Od1Hlk/01/qZyjTuXaiUg5oj5LPPzwwzjvvPNQXl6Oc889F1ptZmbtSB7eLGd9uw0alQCb053QnriJEEtu16hTY2xpLvKztdjdFDrvmuwcZd9srkGnVlw2l/03e6RyjjqVayci5YjoO0FOTo7vp28AcDgcuOCCCyAIArKzs/22FQQB7e3t8lRJaS/ZPXFTQSR512TnKJnNTR2pPFepXDsRKUdEZ4rzzz/f72KYSE7ezGbvbhK5WRqM6NVNItOF8/fm++YoBUHA4Nws7KjvxLqaJswYV5KQi4RwaiVlSOW5SuXaiUgZIvqO+PTTT8epDCJ5euIqoddoPGsIJ++qlBxlsrK5qdCnOpHHaTjvJddcJWP9JeM4i8d+KuHcRZSpol5xd9xxB375y1+itLT/T9+HDh3Ck08+idtvvz2m4igzRZPlTHZGNtE1BBsjpeUoE5XNTYU+1Yk8RqJ5r2jnSgnrLxHHWbD9jMdrJvt4JcoUUZ85li5dirPPPjvgxXBdXR2WLl3Ki2FKmGRnZJVSA5C5OcpU6FOdyGMkXd8rmYLt55kjTbK/ZjqNHZGSRf1jpyRJA36tq6sLGo0m2pcmiogSeo0qoYbeMq2Xbir0qU7kMZKu75VMofaz2+GW/TXTZeyIlC6iW0TffPMNtmzZ4vv3u+++ix07dvhtY7Va8fzzz6OqqkqWAik1JDPv1jsj6/JIsLvc0KnFhGZk5czpBhrLSMd3oPy13eWR7VevSso4pkKf6kRmudP1vZIp1H52WJ2yv2a6jB2R0kW0yl5//XUsXboUQM+CveOOOwJup9frsWLFitirI8VTQt4t36CFQavC9/Wd2NtsgdMtQaMS4JGACWV5CcnIypHTDTSWk6oKAQlYv7s5qvHVqUVsqG2RdX6UMOd9pUKf6kRmudP1vZIp1H6a9JH/NjRTxo5I6SK6GL7iiitw7rnnQpIkTJw4EStWrMC4ceP8ttHpdKiqqoJer5e1UFImJeTdjDo19BoVOqxOONw9f+nM7vJAqxKh16gScmdFjpxuoLHcdrAdEgCDTh3V+MZjfpQw532lQp/qRGa50/W9kinUfmZrVbK/ZrqMHZHSRbTSBg8ejMGDBwMAVq9ejeOPPx5GozEuhZHyKaWfbZfdBavTDZNeDQGAywOYsgAJgNXpRpfdpfi+uoHGssioxUc7GgAAxw7NhVatimh84zE/SpnzQFKhT3Uie+Km63slU7D9tFq6ZH9NIkqMqL9rTZkyBQBgs9mwadMmNDc3o7CwEBMmTEBWVpZsBVJyhJMHTXbezVtjt90Fi8ONCrMRxTk62F09d4cbOu3osLnwXV07xpbmxv0izZvTPWWkGXubLSgvNGCQKby1EGgsHW4JYq//r1X/b3xbux0h9yvW+Ql0DCR7zoORo091vCWyJ+5A79Vld6G+3Sbreyerp3SiBdtPaxxeM96UlPsnSqaYjv77778fd955Jzo6OiBJEgRBQE5ODhYtWoTf//73ctVICRRJHjRZebe+NWZpRLR0OeDyeDA4NwsGnRpujwe1TV2QADy+Zhfys7Vxz7XGkqUNNJZalQDPka4tWlXPhafb7UFtkwUSJDy+djfyj/z6X875cXkkvLrxQMD9SIWMY6L6GscikTV638vp9gw4r3KtiVQYeznEYz8TOXZKzP0TJVPUK++RRx7BDTfcgDPPPBMXX3wxSkpKUF9fj+effx433ngjNBoNrr32WjlrpQSIJA+arLxb3xobO+3otrsgAb46apu60GF1waTXwCMBu5vin2uNJUs70FgWGXWQAOxqtBzZLws6rE6Y9Gp4PFLQ/Yp2flbvaMDr21oG3A9mHFOTErPelBw8Foj8Rf2d68EHH8RPf/pTPPvss36PL1iwAD/96U/x0EMP8WI4xUSTB0103m2gGrcf6oBaJaI0NwsWh9v3oalJFQVQqcS451rlyNIGGsve3SRaux2QIMGkV2NSZSFUYuj9imZ+Nu9rDbofzDimHiVnvSmxeCwQ9Rf1EV9XV4dLLrkk4Nd+9rOf4dVXX426KEqOaPKgici79c61DVRjXrYWWrWIhdNHot3qwONrdsEj9XyIzmJ3xbXvcJe9J5fc2u0IOnYAgo5RsMzxrGMH47u6djy+djc8HgkqUQz4HnLMj8XhRq5eA/cAPZuHFWTHvYcxRa9vDjTc45MXQJmh1eJAq9UJnVqE2yNBrRJ4LESAOevIpcKYRV3VUUcdhcOHDwf82qFDhzBixIioi6LkiCUPGo+8W6Bc2/Hl+TBoVWjstAescUi+HkPy9cjVa7BpXyv2NHXB5QHUYs+F8YSyfNlyrb3ra7U6UdPYBQECBuXooFKJvrqGF2bjk52N2LinNWg+L1iOz6hTY2xpLvL1Guxuiu/86DUiauu6Qo5dPHoYU/T6Hj8GrQp6jQpWpxvtNteAx6dSst4Uf063B5/80Iiahk60dztR327F4Dw9ygqyeSyEwJx15FJpzKK+elm6dCl++9vfYsKECX69hr/55hssXboU999/vywFUuIoreflQLm2Yfl6GHXqoDX29B12xbXvcN/6vH/2d31tCyrMBl9deo0Kq7bWh8znhcrxJWp+stThjR1zh8rSdz6+r+/05csrzMYBj09mvTPHyi11WLWtHgIEiKKAxi4HOqwu1LVZUVZg4LEQBM93kUulMYv6qH/qqafgcrkwfvx4HH300b4P0H377bcoLS3FihUrfH+FThAEvPnmm7IVTfGjlDxosFyb1enGzGNKfHda+9b4v77DGogC4HRLyNX3/EU6ufoOB6pvkEmH9bubIUGCKACVZgOOL8/Hl3taQubzws3xJWJ+bK7QY8fcobL0nQ+XR8LeZgscbg8EAMU5uoDHJ7PemaP3MTKpogD7WrtR325Di8UBCcDMY0p4LAyA57vIpdqYRV3JN998A7VajWHDhqGjowMdHR0AgGHDhgEAtm7d6tvWm1Ej5VNKv9De2eC+2VWLw43TRhbhvPFDAtbYanEc6TtsQFGvvsONnXZYHO6gmbhws02BsssqUUSF2QiVKODKKZUYW5qLVosDa35oDJnVDDevnYj5sTo9IcdOyf2GM1Hf+bC73HC6JejUIlwewO7ywKBT9zs+OUeZo/cxolL1nKuG5mdjf0s3srUqnDaySHG/ulYKnu8il2pjFnUle/bskbEMUppk9wvNN2hh0KrwfX3HgNnVgWrsm33W6NQh85GRZptC5at7X2iEk8OONK8dz/kxaFWobXcGHbtU6DecSfrOh04tQqMSYHd5YMrqyXcPdHxSZgi0ZtVizzEyJE/PNRsEz3eRS7Ux49mQFMmbtY0m9xtNtjbSbFO47yH3dolQXZaPPR0tKVMvBZ4PjwRoVSIkAA2dds5PhuOajR7HLnKpNmYxVWO32/H0009jzZo1aGpqwrJlyzBy5Ei8+eabOOaYY1BZWSlXnZRhYs39RpKtjTbbFO57yL1dvE0bXQxo9ClTL/XoOx8TyvJ83SQsDjfnh7hmY8Cxi1wqjVnUF8NNTU2YNm0avv32W5SUlODw4cPo7OwEALzxxht4//33sWzZMtkKpcwSS+4XiCz7HE62ybtd79cJ9z3k3i7e1KIQU71ddhfq222K7imZjoLNhxJ7fCq1rnCkau1KOcekIo5d5FJpzKKu6sYbb0RbWxu++uorHHvssdBq/5f/mDZtGu69915ZCqTMFE3uN5BwsrXBsk3h9AgON78r93bxFmm9TrcHr248kBI9JdNZ33lTyvHklUq9R/sKVnsqUdoxkUo4dpFLhTGLurq3334b9957LyZMmAC32+33taFDh+LAgQMxF0eZK5F5o2DvFW6PYEqtnpKUPKl8nASr/cyRpiRXR0TRivrH8I6ODpSXlwf8mtPphMvlirooIqAnbzSveggqzQZo1SIqzQbMqx4Sl7swgd5r5rgSXz55dEkOSvP0GF2S45cxph59c9ccKwoklY+TULV3O9yhX4SIFCnq22sVFRX473//i+nTp/f72oYNGzBq1KiYCiNKZN4o0HuF2yNYyRKVbVRiT8lUzXWmMyUeJ+EKVXuH1ZnkCokoWlGfdS655BLce++9GDduHM455xwAPSeGL7/8Eg899BBuvfVW2YqkzJbIvFHf90qlPom9JTqXqaSekqmcSU13SjpOIhWqdpNek+wSiShKUV9h3HTTTVi3bh3mzZuH/Px8AMCMGTPQ3NyMs88+G9ddd51sRRIlQ6r1Sewt0blMJY1VKmdS052SjpNIhao9W6tKdolEFKWozzwajQbvvvsuXnzxRbzzzjs4fPgwzGYzzj33XPzkJz+BKPIODKW+VOqT6JWsvwmvhLFK1r5T+JRwnEQrWO1WS1eSqyOiaMX0XUEQBPzkJz/BT37yE7nqIQopmixotPnRVOqT6JWsXKZ3rE4ZacbeZgvKCw0YZMqS/X2CkWPfUyVrrKQ6I6klFdeUV7DarUmujYiilxpnICJElwWVKz+aCn0SvZKVy1RCVjeWfVdC/eFweSTF9HOOZcxSaU31lcq1E1F/Ea3miooK392WcOzevTvigogGEk0WNBPzo8nKZSphrGPZdyXUH47VOxrw+rYWRdSZKmNGRBRMRLcRpkyZ4vef2+3GwYMHUV5ejv/7v/9DeXk5Dh48CI/Hg6lTp8apZMpE0fQnTeWeprFKZI9mQFljHc2+K6n+UDbva1VEnak0ZkREwUR0i+jpp5/2/f/nnnsO69atw86dO1FWVuZ7fO/evTjzzDMxZcoU2YokiiYLGs1z4pnDTGTGM9G5TCX1j41m35VUfygWh1sRdbZaHGjtdkCnFuHySNCoBMWOWapTUj6cKB1Fvar+9Kc/YenSpX4XwgBQXl6OxYsX46677sKCBQtiLpAIiC4LGslz4pkXTWYWNVHZRiX2j41k35VY/0AMWhVq251JrdPp9uCTnY2oaexCe7cThztsKMnNQll+tiLHLFWlSo6dKNVF/V1y165dyM3NDfi1/Px87NmzJ9qXJuonmixoJM+JZ/YxE3KVqdw/Fkit+qvL8rGnoyWpda7cUodVW+shAFAJAho67WizOlHXZkNZQbbixixVZcK5g0gJoj5bDR8+HP/4xz8wa9asfl978sknUV5eHlNhRH1F0580nOfEszdtJvW9TeX+sUDq1D9tdDGg0Setzt7H9KTKQuxr6cahNitau52QIGHmuBLFjVkqyqRzB1GyRb2S/vjHP+Kyyy7DxIkTcdFFF6GkpAT19fX497//jY0bN2L58uVy1kkUVRY0nOfEMy+aSlnUWKVy/1ggdepXi0JS+zn3PqZVoogKsxHD8rOxr6UbBp0apx1VFJdf4WdabjaTzh1EyRb1Srr00ksBALfddht+//vf+x4fPHgwnnzySfziF7+IuTiiQKLJwQZ7TjzzoqmURZVLqvdgVXr9ye4zHOiYVokC7C4PhuRpZD+mMzU3m4nnDqJkiemMf+mll2LBggX4/vvv0dzcjMLCQowaNSqiXsREyRbPvGgqZVEpNSS7z3Cij+lMzc3y3EGUODGvJkEQMHr0aDlqIUqaeOZFUyWLSqmhd5/hZOVIE3VMZ3pulucOosRI37MIUQTimRdNlSwqpQYl9BlO1DEdTm7Wu12y11W3w+37X5NMr+kd51D58EzLUxPJjauGqJd45kWVnkWl1KCEPsNe8T6mg+Vmhxdm45Odjdi4pzWpWWJvpnnzrjqcqAMe/XgnqqtKE9KjPFPz1ERy43dmIqIUooQ+w4kSLDer16iwamt90rPE3kyz2m3DiSXAgVYr9mw+KEsdofLSmZqnJpIbf3QkIkoh00YXY171EFSaDdCqRVSaDZhXPSRtc6Rzxpf229+Z40pgdbp9WeLSPD1Gl+T4ZYwToff7VZgNAIAKs0GWOvrmpfvu4+EOW9CvJ2oMiNJBet1GoLTA/BvRwLx9hjMlgx4on9xqcWDND41hZ6ejPaeEel4ye5TvbbawDzGRTLhSSDGYfyMKX6Zl0Pvubzg9eKM9p4T7vN6ZZim35/FE9SgvLzSwDzGRTCI6k4qiGHYPYUEQ4HLx1zQUPubfiCgc4fbgjfacEu7zetdR22TB+BKgtskCoy4r7j3KB5my2IeYSCYRrZbbb7+df1CD4iLT+4kSUWRC9eCN9pwS6fO877d5Vx2ANgzN1/u6ScR7H9mHmEgeEV1dLFmyJE5lkBxSOWsrV/YulceAKNUkc72F6nUc7JzSZnXiu7p2jC3N7Vd37+e5PBLsLjd0anHAc5Hd5cHEigJMKM3CNxsOYeH0kSgx5ydkH9nDnEgeXDVpIB2ytqHycaHyb+kwBkSpQknrbaDsdKBzitvjQW1TFyQAj6/Zhfxsbb+68w1aGLQqfF/fib3NFjjdEjQqAR4JmFCWN2AeeVCWByfqAK1a/v0PlQ/PtPw4kdwiWj2ffPJJRC9+2mmnRbQ9AHR2duLOO+/Eli1bsHnzZjQ1NWHx4sUB70o7nU488sgjWLFiBWpqaqDT6TB27Fj89a9/xcknnxzxe6eqdMjahpsBHEg6jAFRqkiF9RbonFLb1IUOqwsmvQYeCdjd1L9ubw/jDqsTDrcHOrUIu8sDrUqEXqMaMI98oNWKE0uA1TsacP6k3KTtNxFFLqKL4alTp4aVGZYkqeencLc74oKam5vxxBNP4LjjjsPcuXOxfPnygNu53W7MmzcPn332GW688UacfPLJsFgs2LhxIywWS8Tvm6rSKWsbbf4tncaASOlSab31Pqe0WZ2QAJj0GkyqKIBKJQasu8vugtXphkmvhgDA5QFMWYAE+Hobe1+z9xhkSSKANmze14oZ1S7FjAERhRbRal29enW86vApLy9Ha2srBEFAU1PTgBfDjzzyCFatWoV169Zh0qRJvsfPOeecuNeoJPHsc5lo0ebflDIGzCtTJlDKegum91r0nlO+q2vH42t2wSMBqiORiEB1t1ocsDjcqDAbUZyjg93lgSgAh9pt6LD1bAcg4BgAgMXhVsQYUGDdDjfaW7p5niY/ER0JU6ZMiVcdPuF2q3jooYdw2mmn+V0IZ6JYs7ZKFGn+LdljoKT8JFG8JXu9BRNsLY4tzUV+tha7m4LX3Xv/Bpl0aOi04VCbFa3dTuRma/DJD42YdezggGMAAAatKiXPu+nO5emZn0c/3onDNpHnafKTkj8W7d+/H3v27MHs2bNxyy234B//+Aeam5sxatQo3HjjjViwYEGyS0yYWLO26SDZY5AK+UkiuSR7vQUTai2GU3fv/Vu/uxmdVhfsbg+0ahECBKzaVo8sjarfawlOK8aXANVl+Rlx3k01q3c0QA3gQKsVkkbP8zT5iWjF3nHHHfjlL3+J0tJS3HHHHUG3FQQBixYtiqm4gRw8eBAA8Mwzz2Do0KF49NFHkZubiyeffBKXXnopHA4HfvWrXw34/IaGBjQ2Nvo9VlNTAwDo6upCR0dHXOqOl2lVOYCzAJv3tcLicGOYUYPqsnxMq8pR1L54s9zxyHQnawy6HW5s3lUHtduG6hJDzx2iXBG1TRZs3lWHyeUGZGtVcXt/OcVzfih2SpofJZ5zwlmL4dY9rSoHDmsuXvyqA1qPG7l6DQqNOgzOzcLe5m5s3lWHK6dW+b1WYU7Pt9OJwwyKOu9Sz7GxfX8DjtECY8xa2AVVyp6n05Xc57eurq6Ithck7+92wiCKItavX4+JEydCFIP/WiHaD9D11tTUhKKion7dJD7//HNMnjwZWq0WP/zwA8rLywH0fHDvhBNOQENDA/bv3z/g6y5ZsgRLly4N+LWHH34YZWVlMdVNRERERMmxb98+XHvttdi2bRuOPvrokNtHdGfY4/EE/P+JVlhYCAAYPXq070IY6LkAnzFjBu655x40NDSguLg44POvuuoqXHDBBX6P1dTUYO7cuZg4cSLGjBkTv+IzmMViwYYNGzDhhBOxYb/Fd0fFoFX13J0ZXQy1mFp/4bDb4cajH+/EgVYrKswGX3awtsmCofl6LJw+MmXuOHjnZ+LEiTAYDMkuh/rg/AQn91oM9HpujwffHGiHBAllBQaYstQ4blgeIAHfH2zEMdombHWYMWZYcb/zmc3pxt/W7MKm/a2wOz3QaURMGJaPK6dWIUuTGueIVNXtcGP56u9wjLYJO1zFsAvalD1Ppyu5z2/bt2+PaPuUDDZVVVUhOzs74Ne8N7qD3bkuLi4e8ELZaDTCZDLFXiQNaMN+C17f1uLL9dW2O7GnowXQ6FMuu2UCUF1Vij2bD2Jzvb1XDjEL1VWlsv0lqkQyGAxcAwrG+QlM7rUY6PVqmyzosHpg0quR61JjX6MTGw4cggRgiAE4pgSoaXFhR3v/89k9r36DVdta/te7uNONPe0tsEKLe84/VtaxIH8mAGOGFQOHm7C9yQFJo0r583S6kuv8ZjQaI9peto9QdnR04KuvvsLevXvleskBqdVqnHfeedi+fTv27Nnje1ySJLz33nuoqqqC2WyOex0Unc37Wn39OUvz9BhdkuPXuzTVzBlfinnVQ1BpNkCrFlFpNmBe9ZCQ/ZGJSF5yr8XerycKgAQJJr0akyoLUZqnx4giAxq77GjqsmNYgR4AUGE29DufHe6w4bNdTXC4PSgx6VBg0KHEpIPD7cFnu5pwuMMm2xhQYNNG99wAG5qv53ma+on4zvA777yDf//739BoNLjkkktwxhlnYNmyZbjxxhthtVoBAPPnz8e//vUvaDSaqIpatWoVLBYLOjs7AQDfffcdXnnlFQDArFmzkJ2djTvvvBOrVq3C2WefjSVLlsBkMmH58uX4+uuv8dJLL0X1vpQYFodb0T1KIxVtf2Qikpfca7H3631X147H1+6GxyNBdeQ3jw63BPHIeczp7vmtZKDz2d5mC2wON3RqEYLg7XEsQqcWYXO4sbfZgkGmrBj3noLxRlYWTh8Jp6jjeZr8RHQkvPPOO5g9ezZ0Oh20Wi3++c9/4tFHH8U111yDs846C8ceeyw2b96M1157DY8//jiuvfbaqIr6zW9+43eH+eWXX8bLL78MAKitrcXw4cNRVVWFTz/9FH/84x9xxRVXwOl0Yvz48Vi5ciXOPffcqN6XEsOgVaG23am4HqWxirQ/MlEs4vlHXlL9D8jEuhb77r9Rp+7pU6zX+PUp1qoEeI5E8zSqnoutQOez8kIDsrQqNHc5IEkeCIIISfLA7vLAaNSivDB9M+BKO5aytSqYTIFjlpS5Ijoy//rXv+L000/HW2+9haysLFx33XX47W9/iyuuuAKPP/64b7vLL78czzzzTNQXw72jD8GMGzcOb7/9dlTvQclTXZaPPR0tiutRSpQKXB4Jr248EJc/8pLpf0Am2P4P1F+5yKiDBGB/ixUnlAC1TRYYdVl+57NBpiycUmXGqm31qO+w92SGXR5oVSJOqTKn5V3hTD+WKLVEdOXx7bff4sknn0RWVs/C/cMf/oBHHnkE8+bN89vu/PPPx+uvvy5flZRWpo0uBjR630my0mzwnSSJKLjVOxr8PoAq5x8PyPQ/IBNq/73nqN7nrklVhYAEfFN7CEAbhubrUV1V2u98tmj2WADAZ7uaYHO4YTRqcUqV2fd4usn0Y4lSS0QXw01NTSgpKfH92/v/+35YrbCwEO3t7TKUR+lILQrM2BJFqfcHUAVBwODcLOyo78S6mibMGFcS9Vrq/aEvuV87FYS7/wOdu06rMOKLdYewcPrIgN0JsrVq3HP+sTjcYcPeZgvKCw1peUcY4LFEqSfio9H7oae+/58oUszYUqaLJk8Zrw+gtloc6DxyFy9dPtwayfi2WhxotTqhU4tweySoVcKA+x/o3OXtUztQv9retUysKIypVqWL17GUTmNEyhLx0fT9999Dre55mvcvzO3YscNvm77/JiKi/4klTxmvD6DmG7TI0amxu9OS8h9ujXR8nW4PPvmhETUNnWjvdqK+3YrBeXqUFWTHvP+haknHbK3cx1I6jhEpS8QXw5deemm/x372s5/5/VuSJN41JiIaQCx5ynh9AHWgD4il4odbIx3flVvqsGpbPQQIEEUBjV0OdFhdqGuzoqzAENP+h6olHbO1ch9L6ThGpCwRHZErVqyIVx1ERBkh1jxlPD+AGugDYqn24dZIx7f39pMqCrCvtRv17Ta0WByQAMw8piTq/Q9VyykjzWmbrZXrWGL+mBIhoiNowYIF8aqDiCgjxJqnjOcHUPv+0QqtWoTD1dMPVym/jg6VG410fHtvr1KJqDAbMTQ/G/tbupGtVeG0kUVR73uoWvY2WxSb0441nyvXH0BJxyw7KQ+PICKiBJIrTxnPD6Dq1CI21LYoKqMZbm400vENtL1aFGB3eTAkTx/XHHZ5oUFxOW2587mxHqfplGUn5eLFMBFRAqVCNleJGc1wa4p0fOM5H6Fee5ApS3HHgtLmPhXWC6U+HkVERAmm5GyuEjOakdYU6fjGcz5CvbaSjgUlzj2grDGi9MSL4SRhv0SizCVXnjIe5MxoynWei7SmSMc3nvMR6rWVdCwoNZ+rpDGi9MSjKcHYL5GIvJT4h2fkyGjKfZ6LtqZIxzee8xHqtZVwLCg9n6uEMaL0xKMqwZSWxyIi6k2OjKbc5znmRhOD40yZikd2Aik1j0VE1FssGc14neeYG00MjjNlIl55JZBS81hEpHzdDrfvf01xfq9wM5qBMsHxOs8xN5oYHGfKRDzCE0jpeSwiUh5v/nbzrjqcqAMe/XgnqqtKE/I5g4EymsEywfE+zzE3mhgcZ8okPNITiHksIoqUN3+rdttwYglwoNWKPZsPAlBuz1+e54golbB9QYLNGV+KedVDUGk2QKsWUWk2YF71EOaxiKif3vnbCrMBAFBhNvg9nsyaRpfkoDRPj9ElOX6P8zxHRKmEP6InGPNYRBQuJX7OoNXiQKvVCZ1ahNsjQa0S+tU0rCDb7zynVYtwuDywuzxp20IyVKabveWJlIsrMkmYxyKiUHrnb6XcnovIZH7OwOn24JMfGlHT0In2bifq260YnKdHWUF2wJp0ahEbalvSuq96qEw3e8sTKR+vxoiIFKr35wxqmywYXwLUNllg1GUlJX+7cksdVm2rhwABoiigscuBDqsLdW1WlBUY+tWUCX3VQ2W6M2EMiFIdfywlIlIwb/52aL4eADA0X5+U/G3vTPCkigKMLslBcY4ObkmCBGDmMSV+NYWTLU51oTLdhztsaT8GROmAd4aJiMKQrMyn93MGk8sN+GLdISycPhIl5vyEvb9X7/yySiWiwmzE0Pxs7G/pRrZWhdNGFvn92l+JeeeBRDu3vffRI0kAAI8k+fZxb7Ol3xi4PBJ0ahGt3Q5FjQFRJuMqJCIKQimZz2ytyu9/Ey1Q/2C1KMDu8mBInr5ffjkV+qrHOrf5Bi0MWhW+r+9AZ4cbEyqArQfa0ORQYUJZPsoLDb4xGJSjw77WbtS329BicSA3W4NPdjbixycMY3aYKMl4MUxEFAQznz0i7ZOeCn3VY51bo04NvUaFDqsLoqsn8tDS7USHU4Jeo8IgU5ZvDNbXtqDD5oTD5YFOJUIAsGprPbLUqow6joiUiD+OEhENIBNyr5GItH+wkvsNyzG3XXYXrE43THoNCo/c6S40aGHSa2B1un09l2eOK4EECR6PhCKjFqMG52BSZWHGHkdESpP8H82JiBQqlXKv0YokLxtpn3Q5+6pHk+sN9hw55rbV4oDF4UaF2YBykwDAgqOH5GJvhwSLw+3ruXzaUUX4cEcDuu0ulBVkQ30kFpFOxxFRKuPqIyIaQCrkXqMVS1420j7psfRVj6bOcJ4jx9z2fg1Vrg4AoBIEtFsdfq+Rb9AiX69Bq8UBldhz4Z0uxxFROuDFMBHRAFIh9xqtVMlCR1NnOM+RY27D7QOdzscRUTrgCiQiCsKbb/XeZaw0G3x3GVNV37ysIAgYnJuFHfWdWFfThBnjShRxgRZNnZE8R4659W67eVcdgDYMzdf7/gJdoO3S6TgiShfJP9sRESmYnLnXRAgnWxtuXjaevZX7vrb331q1CIfLg3yDNqJcr/f53XZX2M+xuzyYWFGAU0aafe8ZaD+DjUPfPtCXn1oJtS4bdpfHL8Yx0HHUZXehvt2m+OOKKJ1x5RERhSGW3GsiRJKtDZWXNWap8erGA3Hprdy3ToNWBb1GBYvDhV2NFnTYnDBlaTCy2IgThxfAoFWhsdM+YK637+tlaUS0dDng8njCfk7v/Yt2TLXqnn//49PdOGwTB9zWexw53Z64jTERRUa5Z3YiIgpbJNnaUBnWj7c3xC1P3LfO7+s7evr0ioDkAexuDzptLnTaXGjqcmBYvh5GnXrArO2rGw/4vV5jpx3ddhckIOznDLR/kYzp6h0NUAM40GqFpNGHHLNUyWwTZQL++ElElOKi6Zk7UA/g6WOK49ZbuW+dxTk6COiJK9idHggCUGLSQSUIEAWg3eqE1enGzGNKAvYqHmi/9VoVTHoNhuXrw35O3/2LZEy77C5s3tcKAKgwG0KOGftXEykL7wwTEaW4aHrmBsqwAsB3de1otTrj0lu5b512lxsuD6BVC7A7PVCpBAiCCI1KhNMtIVurgsXhxmkji3De+CH9crv17baA+52XrYVWLWLh9JHI1qrCek7f/YtkTL39hqFDWGOWCf2riVIJ7wwTEaU4bwa43eqEJEkA/tfHNkenDtrH1qhToyQ3C+9vq8ftb2zD42t2oaahE7VNXXB7PBG9VqR16tQi1CLgcElQiQLcbgmS5IHT7YFGJaDb4fa9p1GnxrCCbL+LxFD7PSRfH/FzevcGDndM8w1aGLQq3zahxiyW+SIi+fFHTyKiFBdrH9u++VUBAjqsTqzf3YwKs1G2nriB6pQA6NSiLzNc32GHVi3CI/X8hbZg7xnNfof7nEhe26hTo7osHzh8CLVNFkgaT9A62HeYSFm44oiI0kC0fWwD9eUdlKPD+toWSOi5YytnT9y+dU4oy/d1k9jdaEF7r24Sp44sCvme0ex3uM+J5LWnjS7Gp4e/w9B8PQ7bxJB1sO8wkXLwYpiIKAX17X0bbT/kQPlVlUpEhdkAUQCunFKJ8kIDHC5Pv9650QjWb7dvn+Fw6o9mv73POWWkGXubLSgvNGCQKSum11Yf+TPLC6ePhFPUhawj1fpXE6UzrjwiohTi8khB+9NG2g85WM/h4YXZ2NnQhRc27Je9F27fOmPt4xzJ8yPpHxzpa2drVTCZsuNSNxHFB1cgEVEKWb2jAa9va5GtP22w/Kpeo8KqrfVp1wuXPX6JqDd2kyAiSiGb97XK3p82UM/hmeNKYHW6064XLnv8ElFfvDNMRIrTNw+brqLZT4vDLXt/2kD51VaLA2t+aEy7Xrjs8UtEfXHFE5FiRJrlTFWx7KdBq0Jtu7NfvrfSbIi5P23f/OpAWWI53itZgmWkU3m/iCh6vBgmIsXIlCxnLPtZXZaPPR0tce9Pm669cNN1v4goelz1RKQIgfrdDs7Nwo76TqyracKMcSVpcaES635OG10MaPQJ6U+brr1w03W/iCg6qf+dhYjSQqZkOWPdT7UoDNint77dNmD+OJp8cri9cCN5bbnz4PHcLyLKDFz9RKQImZLllGs/vflep9sTtO+wHDnsgXrhRvLacufB47lfRJRZeBYgIkXIlCyn3PsZKn8czxx2JK8tdx2Zki8novhLn49nE1HKC9Tvdl71kLTLcsq1n6F65h7usMWtp24k/Xrl7u3LXsFEJKf0uNVCRGlBCVnORPQ4jmU/my0O7GhuRnmhAQ6Xp1/+2OWRoFOLaO12YG+zJaJ88kD7HujxSLLPwbZtszrxXV07xpbmhj0GcuXLM6WfNREFx9VPRIqTjCxnMnocR7KfNqcbAPDbFzfjYBeQpVXh/yoKoNeIaOy0Y1CODvtau1HfbkOLxYHcbA2+reuAQatCY6c9aD55oH2feUwJVm2tDzgmkWSfA23r9nhQ29QFCcDja3YhP1sb9njHmrvOlH7WRBQeXgwTEUH5GdS/rdmF49RAW7cTLo8azV0OfPhdA0aX5MCoU2N9bQs6bE44XB7oVCIEAB9tb8CwfD2MOnXQfPJA+/7Vnhbsb7UOOCbhZp8D5aRrm7rQYXXBpNfAIwG7m8If71hz10qfayJKLP4ITEQZT+kZ1MMdNmza3woAMBu1KDDoUGLSweH24GCbFZOrCiBBgscjocioxajBOZhUWYguuwtWpxszx5UMmE8eaN/brU58tqsJHVbngGMSSfa597YqUYAEwKTXYFJFQVTjHW3uWulzTUSJxzvDRJTxgmVQW7sdEWda5ba32QK70wMAkCQBTrcHKkGATi3C5nRjWGE2RhTnoNvuQllBNtRHftWfq9fA4nDjtKOKcF71kID52N777vJIsLvc0KlFZGtVqG93Q5+r8nu8dy53WEF22Nnn3jnp7+ra8fiaXfBIgOpIrZFmfr2vd8pIM/Y2W1BeaMAgU1bIscyUftZEFD6ueCLKeAEzrW4PapsskCDh8bW7ka/XJC1XWl5ogFbdc+HWarGjySZCBOBwe1Bs0mHs4Fx8rG9Eq8UBldizXd8M7UD55HyDFgatCt/Xd2JvswVOtwSNSoDLI0GrFlDXZsXhDpvvcY8ETCjL88vlRpJ9NurUGFuai/xsLXY3Rd9rOdrcb6b0syai8PFimIgyXuBMqwUdVidMejU8HimiTKvcBpmyMMikB2CFzeWBwwV4JAmiIGBoXjYqi4xRZ2iNOjX0GhU6rE443B7o1CLsLg+0KhGFBi2aLY5+j+s1qpjunsrRazna3G+m9LMmovBx1RMRAb6s6bqaJrR2OyBBgkmvxqTKQqhEEYNzs7CjvhPrapowY1xJQi+auuwuFB65Y6lVC1CLAkRRgF6jRrFJ58vveuvvtLtQaTb47pSGem2r0w2TXg0BgMsDmLIAtwTY3R4Ys1RQCWrf4xIAq9ONLrsrpjGItl5vzb1zv4IgRDQ/sbw3EaUfXgwTESFApnXtbng8ElRidJnWcITb57bV4oDN1ZMZPrnKjAabCKNOjdZuJ6xOT8T53b6vbXG4UWE2ojhHB7ur5y7wvpZu7G7sQmWREWUF2b7HGzrtsDjcaLU4fM+Ppk9vtL2Wu+wufFfXjlarM+rcbyL6WbOHMVHq4AolIurFl2nVa2LKtAYTad7Vm+sFAJ1ahaIc3ZF6uvvVE2mP5r4ZWoNODUmSYHW4kaVVodvhhkoUfI+3W50YXpiNT3Y2YuOe1pj79IZbb+8xa+12oKaxCwKAQSYdVKIY1fzEo5+1yyPh1Y0H2MOYKIXwYpiIqI9450ojzbsadWpUl+UDhw/1fKhP45GtnoH21aTX4OhSE/a3WvuNgV6jwqqt9Qnt09t3zAQI6LA6sX53MyrMRsXkflfvaMDr21rYw5gohfDHVCKiAKLtYxtKtH1up40uBgAMzdfLWg8w8L4umj223+Mzx5X4MsOJ6tMbaMwmVRTApNdAAqASBVnHIxab97WyhzFRiuGdYSLKKOFmOfvmSrVqEQ6XB3aXJ6Zfd0fb51Z9pGXawukj4RR1EWdRg+13sAxt3zHY22zBhzsaYu7TG0mmNtCYqVQiKswGiAJw5ZTKpPaB7s3icMe9hzHzyETy4ioioowQbV9anVrEhtoW2TKgsfa5zdaqYDJlh/1+kez3QBna3mMQa143mnkINWZKuRAGAINWhdp2pyKy5kQUHmWcPYiI4izavrTRPm8gie5zK0f9cuZ1o6knlXoDV5flY09HiyKy5kQUHv4oSURpL9qcbrTPCyVeeeR41C9nXjeWehI1ZrGaNrpYUVlzIgpNOT9OExHFSbQ53WifF0oi+twCkdcfKIsqZ143lvEMNWaBak9GtlYtCnGZ23gdi0TEi2EiygDR5nRjzfeGEo8+t72FW3+wLKqceV05xrPvmAWqfVJVISAB63c3Jy1bK/fcxvtYJMpkiotJdHZ24sYbb8RZZ52FoqIiCIKAJUuWBH2OJEk47bTTIAgCFi5cmJhCiShl9M5s7qjvRF1bT+/cUFnOaJ+nFOHW782i7m6ywOHyYHeTBa9vPoiVW+pkHYN4jGeg2pd/shtPfro74P6kqlQ/FomUTHGrp7m5GU888QSOO+44zJ07F8uXLw/5nMceeww1NTUJqI6IUpU3s+m9g1hpNvjuFsbjeUoRqv6+WVRBEDA4Nws76juxrqYJM8aVyDoGcr5WoNqLjFp8tKMBAHDs0Fxo1ap++5OqF46pfiwSKZXizgjl5eVobW2FIAhoamoKeTG8Z88e3HzzzXj22Wcxf/78BFVJRKkm0pxu77xpIvK98RJqv8PJog4ryJZtDLz1nDLSjL3NFpQXGjDIlBXVawWq3eGWIPb6/1p1+mRrE5U1J8o0iltF3hNauK644gqceeaZmDdvXpwqIqJ0EirLGSw/O6wg/P6+SjPQfkeSRZUjBytnr9xAtWtVAjySBADQqnq+n6RbtjbeWXOiTJPSq2n58uXYsGEDvvvuu2SXQkRpItN6uaZi3+NQtRcZdZAA7Gq0KLonMREpQ8qeFQ4ePIgbbrgBf/7zn1FaGlleqqGhAY2NjX6PeTPHXV1d6OjokK1O+h+LxeL3v6QsnB+g2+HG5l11ULttqC4x9NwlzRVR22TB5l11mFxuQLZWlZTa4jk/06pyAGcBNu9rhcXhxjCjBtVl+ZhWlSPr+TAe4xuo9uOGFQES8PWBtrjuT29cP8rG+VE2ueenq6srou0FSTry+yQFampqQlFRERYvXtyvo8Ts2bPR0dGBNWvW+OXcrr76ajz66KNBX3fJkiVYunRpwK89/PDDKCsrk6V+IiIiIkqsffv24dprr8W2bdtw9NFHh9w+Je8Mv/LKK3jvvffw2Wefob293e9rDocDbW1tMBgM0Gg0AZ9/1VVX4YILLvB7rKamBnPnzsXEiRMxZsyYuNWeySwWCzZs2ICJEyfCYDAkuxzqg/PTc+fy0Y934kCrFRVmgy8/W9tkQWleFo4Zkotv6zpgcbhh0Kpw3NA8QAC+3t/me6y6LB/TRhdDLUb2+YdQQs2PyyNh9Y4G3x1SvUZElloFm8sNq9MT19rCFWx8h+brsXD6yKTdeY8V14+ycX6UTe752b59e0Tbp+TF8LZt2+ByuTBp0qR+X3vyySfx5JNP4vXXX8fcuXMDPr+4uBjFxcUBv2Y0GmEymeQsl/owGAwcYwXL5PkxAaiuKsWezQexud7eK2+aBUmtxzs72n1Z19p2J744cAgCgGyd2vfYno4WQKOPW754oPl5deMBvL6t5X/11VnQYXXCpFejwmxMSG2hBBvf6qpSlJjzk1KXnDJ5/aQCzo+yyTU/RqMxou1T8mL40ksvxdSpU/s9Pm3aNMydOxfXXXcdxo0bl/jCiCjlBerlevzwfHxZ2+LXz9Zs1OLjI/1sTx+aC00S+9n27bfr8kjY22yBw+2BAKA4R6eYXrvslUtESqPIi+FVq1bBYrGgs7MTAPDdd9/hlVdeAQDMmjULw4cPx/DhwwM+d8iQIQEvlImIwhGol2urxYE13zcG7WeriVM/W2+/Y43HHfBxrVrE3mYLWrsdvvrsLjecbgk6tQiXB7C7PDAcuXudjF67vXs2G3Vq9solIkVR5BnoN7/5Dfbu3ev798svv4yXX34ZAFBbWzvghTARkVz69nJNdD/bvv14B2V5cKIOsDnd+GDjAXy6sxE7G7rQYetpG9Zpc0IURAzK0UGnFqFRCbC7PDBlATq1mJReu8F6CrNXLhEphSLPRHv27InqeQpujEFEKWygfrZmow4CgJo49LPt24/3QKsVJ5YAf1uzCz+0urGvxYJOqwt2twedNhdEAfB43Fhf24IKswEeCdCqREgAGjrtSem1m2k9m4koNSnyYpiISGkCZV0nVRYCArB+V7Os+de+GWBBEJAliQDasGl/K5xCFgQAoiigxKBDu9UFo04Nt+SBBAmiAEwoy4Neo4LV6YbF4U54NjfQPiglt0xE1BvPREQkq7750GTrna11uDxR1xUoS+x9nVnHDJZ1n1stDnQeuZvau486ANidHuj0AlweQKsWAQhQCQKcbg+GFWRDpxFx/vFDcOLwQgwyZSVtPlotDrR2O47kliVoVEJcMtVERLHimYiIZBEsH6pRiUmrp3e21pSlwchiI04dWRR1XYGyrnLnX/MN2n4ZZW8MTKcR4XRLUItAs8WJDqsDDrcErUqE7XAnDDo1Xt14EB9vb/SN/7CCbNlqC4fT7cEnOxtR09iF9m4nDnfYUJKbhbL87ITnlomIQuHFMBHJQmn5UG89+1q60WFzwuHyoMvqQqfNiaYuR9LqCkegjLLgtGJ8CTBhWD5+aHWjvsMGm9MDj9TT1cLmdMPu8gCCAI8E7G5K3viv3FKHVVvrIQBQCQIaOu1oszpR12ZDWUF2QnPLREShJP52DRGlnb750NI8PUaX5Pg9nox62q1OiELPBVmJSQdRFCAA6LA6k1JXJOaML8W86iGoNBugVYsYmq8HAFw5tQozjylBtlYFjUqAXqOCUaeGTi1CFIAcnQpFObqkjX/v95xUWYhRg3NQZNTC45EgQcLMcSXsKUxEisIfzYkoZgNlXJOVD/XWk61VocXigEYlQhBEaI/03dVrVXGtS46cbt+MssZjxxfrDiFLo8JpI4vw4XeHUZqnR0luFix2F76t64DN6fb1FdYkqa9w72NBJYqoMBsxLD8b+1q6YdCpcdpRRUmJzRARDYQXw0QUs4EyrsnKh3rrqW+3QaPqiRBIkgcOlwdGnQpWhxtD8vSy1xWP3LQ3j9zR8b+7u/kGLfKztWjttkCvUUGrEqFVCWi3epCrF5LWV9hbW99jQSX29DwekqdhVpiIFIcXw0QUs4H68Ca6r22getqtTrglCfUdduiO9N016TVxqStRuelA4y2hp6+wRwIak9RXeKDaknksEBGFwrMSEckiUB/eRPa1Haie3t0kcrM0GNGrm4ScEt1Xt+94TyjLT2pf4WC1JftYICIKhhfDRCSLYH145RJJFrdvPbH2GQ4l0bnpgcY7Hn2FI33NRBwLRERy4dmJiGQld89dILYsbjzqCSRZuem++yfn/saagU7U2BMRxYJnKSJSPKX1MA4kHbOyqTDuRESxYn8bIlI0pfUwDqZvb+BKswHzqoekZFY2lcadiCgWqXergogySrJ6GEeTvU2nrKzSekcTEcULz2REpGiJzuLK0Ss4HbKySusdTUQUL6l9tiaitJfoLC5zsj3SMQNNRBQIz2ZEpHiJ6lub6F7BSsd+wUSUCTLnrE5EKStRWVwl5mS92WWNx53Q9wXSKwNNRDQQntWIKGXEO4urpJxs3+zyoCwPTtQBLo+UsBq80iEDTUQ0ELZWIyI6oncedkd9J+rarNhR35mUnKw3u7y7yQKHy4MDrVYAwOodDQmrgYgoE/BHfSKiXpSQkw2UXc6SRABt2LyvFTOqXbxTS0QkE55NiYh6UUJOdqDsMgBYHG5Zs8vR9FMmIkonPPMREQWQzJzsQNllADBoVbJkl+Xop0xElA54MUxEpDCBevwKTivGlwDVZfmyXKSznzIRUQ/++E9EpEBzxpdiXvUQVJoN0KpFDM3XAwCmjS6O+bX7ZpJL8/QYXZLj9zgRUabgnWEiykhKz8r2zS5rPHZ8se4Q1KIQ82srsZ8yEVGy8GxHRBkl1bKy3uxyR4d8d2uV1E+ZiCjZeDFMRBmFWdnAmeR2qzMp/ZSJiJKNZzwiyhiB+vcOzs3CjvpOrKtpwoxxJRlzIaiEfspEREqQGWd9IiIwK9ubEvopExEpgfICckREceLNyrZbnb6+vd6sbI5OnZFZWaNOjWEF2bwQJqKMxbMfEWUMZmWJiKgvnvmJKKMwK0tERL3xYpiIMgqzskRE1Bu/AxBRRvL27yUioszGD9ARERERUcbixTARERERZSxeDBMRERFRxuLFMBERERFlLF4MExEREVHG4sUwEREREWUsXgwTERERUcbixTARERERZSxeDBMRERFRxuLFMBERERFlLP4t0iPsdjsAoKamJsmVpK+uri7s27cP27dvh9FoTHY51AfnR9k4P8rG+VE2zo+yyT0/3ms577VdKLwYPmL//v0AgLlz5ya3ECIiIiKK2f79+zFhwoSQ2wmSJEkJqEfx2trasHbtWgwbNgw6nS7Z5aSlmpoazJ07F2+88QZGjBiR7HKoD86PsnF+lI3zo2ycH2WTe37sdjv279+PKVOmIC8vL+T2vDN8RF5eHs4777xkl5ERRowYgaOPPjrZZdAAOD/KxvlRNs6PsnF+lE3O+QnnjrAXP0BHRERERBmLF8NERERElLF4MUxEREREGYsXw5QwRUVFWLx4MYqKipJdCgXA+VE2zo+ycX6UjfOjbMmeH3aTICIiIqKMxTvDRERERJSxeDFMRERERBmLF8NERERElLF4MUyy2LJlC8455xyUlZVBr9ejoKAAJ510Ev75z3/6bXfppZdCEIR+/40ePTpJlWeu5cuXQxCEgH8HftOmTTjjjDNgNBqRl5eH+fPnY/fu3UmoMnMNND9cQ4m3Zs2agGMuCALWr1/vty3XTuKFOz9cO8n12WefYdasWcjPz4der8fIkSNx5513+m2TrPXDv0BHsmhra8OwYcNw0UUXYciQIbBYLHj++efxs5/9DHv27MFtt93m21av1+Pjjz/2e75er090yRnt4MGDuOGGG1BaWor29na/r+3YsQNTp07F+PHj8dJLL8Fms+H222/Hqaeeii1btvDT2AkQbH4ArqFkufvuuzFt2jS/x8aNG+f7/1w7yRVqfgCunWT517/+hZ/97Gf48Y9/jGeffRZGoxG7du1CXV2db5ukrh+JKI7+7//+Txo2bJjv3wsWLJAMBkMSKyJJkqRzzz1Xmj17dsD5uOCCCySz2Sy1t7f7HtuzZ4+k0WikG2+8MdGlZqRg88M1lHirV6+WAEgvv/xy0O24dpIj3Pnh2kmOAwcOSAaDQfrNb34TdLtkrh/GJCiuzGYz1Gr+AkJJ/vnPf2Lt2rVYtmxZv6+5XC68/fbbOP/882EymXyPl5eXY9q0aXj99dcTWWpGCjY/pFxcO0SBLV++HBaLBTfddNOA2yR7/fBimGTl8XjgcrnQ2NiIZcuW4f333++3AKxWK0pKSqBSqTB06FAsXLgQLS0tSao4szQ0NOD666/Hn/70JwwdOrTf13ft2gWr1Ypjjz2239eOPfZY1NTUwGazJaLUjBRqfry4hpLj6quvhlqthslkwowZM/DZZ5/5vsa1k3zB5seLayfxPvnkExQUFGDHjh0YP3481Go1iouLceWVV6KjowNA8tcPb9mRrK666ir8/e9/BwBotVo8/PDD+PWvf+37+nHHHYfjjjvOl+Nau3YtHnjgAXz00Uf48ssvA36Yi+Rz1VVXYdSoUfjNb34T8OvNzc0AgIKCgn5fKygogCRJaG1txeDBg+NaZ6YKNT8A11Ay5Obm4rrrrsPUqVNRWFiImpoa/OUvf8HUqVPxzjvvYMaMGVw7SRTO/ABcO8ly8OBBdHd344ILLsDNN9+MBx98EF9++SUWL16Mbdu24dNPP03++olrCIMyzt69e6Uvv/xSeuedd6Qrr7xSEkVR+stf/hL0Oa+88ooEQLr//vsTVGVmeuWVVyStVit9++23vsf6ZujWrVsnAZBeeOGFfs+/++67JQDSoUOHElJvpglnfoI9l2sosVpbW6WhQ4dKxx57rCRJXDtK03d+BsK1E38jR46UAEj33HOP3+MPPvigBED64IMPkr5+GJMgWZWVleGEE07ArFmz8Pjjj+OKK67AzTffjMbGxgGfM2/ePBgMhn4tikg+XV1duPrqq3HNNdegtLQUbW1taGtrg8PhANDTDcRisaCwsBDA/+4Q99bS0gJBEJCXl5fI0jNCuPMzEK6hxMvLy8O5556Lb775BlarlWtHYfrOz0C4duLPuza8d+i9Zs6cCaCnnVqy1w8vhimuJk6cCJfLFbJPoCRJEEUejvHS1NSEw4cP47777kN+fr7vv3//+9+wWCzIz8/HJZdcgqqqKuj1emzdurXfa2zduhUjRoxAVlZWEvYgvYU7P8FwDSWeJEkAAEEQuHYUqPf8hNqOayd+AuWAgf/NjyiKSV8/nH2Kq9WrV0MURVRWVg64zSuvvILu7m5MmjQpgZVllpKSEqxevbrffzNmzEBWVhZWr16Nu+66C2q1GrNnz8Zrr72Gzs5O3/P37duH1atXY/78+Unci/QV7vwMhGso8VpbW/H2229j/PjxyMrK4tpRmL7zMxCunfg7//zzAQCrVq3ye/zdd98FAEyaNCnp60eQvJfmRDG44oorYDKZMHHiRAwaNAhNTU14+eWX8eKLL+IPf/gD/vznP2Pv3r24+OKL8ZOf/AQjRoyAIAhYu3YtHnzwQVRVVeGLL76AwWBI9q5klEsvvRSvvPIKurq6fI/t2LEDJ554IiZMmIA//vGPvsbnLS0t/MMBCdZ3friGkuPiiy/2RcDMZjN27tyJ++67D7t27cKqVatwxhlnAODaSZZw5odrJ7nmzJmD//znP7jtttswadIkfPXVV1i6dCnOOOMMvPXWWwCSvH7ilkamjPLUU09Jp556qmQ2myW1Wi3l5eVJU6ZMkZ577jnfNi0tLdK8efOk4cOHS3q9XtJqtdLIkSOlG2+8UWpra0ti9ZlroA9offXVV9Lpp58uZWdnSyaTSZo7d65UU1OThAozW9/54RpKjnvuuUcaP368lJubK6lUKqmoqEiaN2+etGHDhn7bcu0kXjjzw7WTXN3d3dJNN90kDRs2TFKr1VJZWZl08803SzabzW+7ZK0f3hkmIiIioozFzDARERERZSxeDBMRERFRxuLFMBERERFlLF4MExEREVHG4sUwEREREWUsXgwTERERUcbixTARERERZSxeDBMRERFRxuLFMBERERFlLF4ME1FGePrppyEIgt9/RUVFmDp1Kt5+++1+2wuCgCVLlvj+vWbNGgiCgDVr1vgeW7JkCQRBCPne4W6XKO+++67fvvUmCAIWLlwY0+t/+umn0Ol02Lt3b0yvI6ePPvoIRqMRBw8eTHYpRKQwvBgmooyyYsUK/Pe//8Xnn3+OJ554AiqVCrNnz8Zbb73lt91///tf/PKXv0xSlfH17rvvYunSpXF5bUmScP311+NXv/oVysvL4/Ie0Tj99NMxceJE3HLLLckuhYgUhhfDRJRRxo0bh0mTJuGkk07CvHnz8Pbbb0On0+Hf//6333aTJk3C0KFDk1Rl6nrvvfewadMmXHPNNckupZ+rr74azz//PPbv35/sUohIQXgxTEQZLSsrC1qtFhqNxu/xvjGJeHjxxRdx0kknwWAwwGg0YsaMGdi8ebPfNpdeeimMRiNqamowa9YsGI1GDBs2DL///e9ht9v9tj1w4AB+9KMfIScnB3l5ebjkkkvw5ZdfQhAEPP30077Xe+yxx3z76P1vz549fq/13HPPYcyYMcjOzsZxxx0XMEoSyOOPP44TTzwRo0aN8nt8+PDhOPfcc/H222+juroaer0eY8aM8b3u008/jTFjxsBgMGDixIn46quvAo7Djh07MGPGDBgMBgwePBh/+tOfAADr16/HKaecAoPBgKOOOgrPPPNMv9pmz54No9GIJ598Mqx9IaLMwIthIsoobrcbLpcLTqcTBw4cwPXXXw+LxYKLL744oXXcfffduOiiizB27Fi89NJLeO6559DZ2YlTTz0V3333nd+2TqcTc+bMwemnn44333wTl112GR544AHce++9vm0sFgumTZuG1atX495778VLL72EQYMG4cILL/R7rUWLFuFHP/oRgJ4oiPe/wYMH+7Z555138Oijj+KOO+7Aq6++ioKCAsybNw+7d+8Ouk8OhwMffvghpk2bFvDrX3/9NW6++WbcdNNNeO2115Cbm4v58+dj8eLFWL58Oe6++248//zzaG9vx7nnngur1dpvHObPn49zzjkHb775JmbOnImbb74Zt9xyCxYsWIDLLrsMr7/+OkaNGoVLL70UGzdu9Hu+VqvFySefjHfeeSfofhBRhpGIiDLAihUrJAD9/tPpdNKyZcv6bQ9AWrx4se/fq1evlgBIq1ev9j22ePFiKZzTaN/t9u3bJ6nVaumaa67x266zs1MqKSmRfvzjH/seW7BggQRAeumll/y2nTVrljRq1Cjfvx977DEJgLRq1Sq/7X79619LAKQVK1b4Hrv66qsHrBuANGjQIKmjo8P3WH19vSSKonTPPfcE3c8vvvhCAiC98MIL/b5WXl4u6fV66cCBA77HtmzZIgGQBg8eLFksFt/jb7zxhgRAWrlyZb9xePXVV32POZ1OqaioSAIgbdq0yfd4c3OzpFKppN/97nf96rj11lslURSlrq6uoPtCRJmDd4aJKKM8++yz+PLLL/Hll19i1apVWLBgAa6++mo8+uijCavh/fffh8vlws9//nO4XC7ff1lZWZgyZYpfxwqgJ84we/Zsv8eOPfZYv24Na9euRU5ODs4++2y/7S666KKI65s2bRpycnJ8/x40aBCKi4tDdoeoq6sDABQXFwf8+vjx4zFkyBDfv8eMGQMAmDp1KrKzs/s93vf9BEHArFmzfP9Wq9UYMWIEBg8ejOrqat/jBQUFA9ZbXFwMj8eD+vr6oPtCRJlDnewCiIgSacyYMTjhhBN8/z777LOxd+9e3HjjjfjpT3+KvLy8uNdw+PBhAMCJJ54Y8Oui6H+fIjs7G1lZWX6P6XQ62Gw237+bm5sxaNCgfq8V6LFQCgsL+z2m0+n6xRb68n69b61eBQUFfv/WarVBH++9f0DgcdBqtf2e73287/N71xZqX4goc/BimIgy3rHHHov3338fP/zwAyZOnBj39zObzQCAV155Rbb2Y4WFhdiwYUO/xxN5B9S7Xy0tLQl7z0h5a/PWSkTEi2EiynhbtmwBABQVFSXk/WbMmAG1Wo1du3bh/PPPl+U1p0yZgpdeegmrVq3CzJkzfY+/8MIL/bbV6XQAeu6O6vV6Wd4f+F+8YdeuXbK9ptx2796NwsLCqO6YE1F64sUwEWWUbdu2weVyAeiJFrz22mv44IMPMG/ePFRUVCSkhuHDh+OOO+7Arbfeit27d+Pss89Gfn4+Dh8+jA0bNsBgMET8RzEWLFiABx54AD/96U9x1113YcSIEVi1ahXef/99AP7Ri2OOOQYAcO+992LmzJlQqVQ49thjffGEaA0dOhSVlZVYv349rr322pheK17Wr1+PKVOmKOovAhJRcvFimIgyyi9+8Qvf/8/NzUVFRQXuv/9+XHXVVQmt4+abb8bYsWPx0EMP4d///jfsdjtKSkpw4okn4sorr4z49QwGAz7++GNcf/31uPHGGyEIAs466ywsW7YMs2bN8stCX3zxxVi3bh2WLVuGO+64A5Ikoba2FsOHD495vy655BI8+uijsNvtvjvQSrFr1y5s3bo17v2jiSi1CJIkSckugoiI4uPuu+/Gbbfdhn379iXkL+rV1dWhoqICzz77bL8ex8m2aNEiPPvss9i1axfUat4LIqIevBgmIkoT3vZwo0ePhtPpxMcff4yHH34YF154IZ599tmE1XHTTTdh1apV2LJlS7/OGMnS1taGyspKPPLII7jkkkuSXQ4RKQh/NCYiShPZ2dl44IEHsGfPHtjtdpSVleGmm27CbbfdltA6brvtNmRnZ+PgwYMYNmxYQt97ILW1tbj55psT/pcGiUj5eGeYiIiIiDKWMn5/RURERESUBLwYJiIiIqKMxYthIiIiIspYvBgmIiIioozFi2EiIiIiyli8GCYiIiKijMWLYSIiIiLKWLwYJiIiIqKMxYthIiIiIspYvBgmIiIioozFi2EiIiIiylj/Hz5xOyfT4c4DAAAAAElFTkSuQmCC", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "x = penguins_clean['bill_length_mm'].to_list()\n", + "y = penguins_clean['bill_depth_mm'].to_list()\n", + "\n", + "plt.figure(figsize=(6,4))\n", + "plt.scatter(x, y, s=12, alpha=0.6)\n", + "plt.xlabel('Bill length (mm)')\n", + "plt.ylabel('Bill depth (mm)')\n", + "plt.title('Penguins: bill length vs bill depth')\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "id": "286f536a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (1, 2)
corr_bill_len_depthcorr_flipper_mass
f64f64
-0.2350530.871202
" ], - "source": [ - "mass_by_species = (\n", - " penguins_clean\n", - " .group_by('species')\n", - " .agg([\n", - " pl.len().alias('n'),\n", - " pl.col('body_mass_g').mean().alias('mean_mass'),\n", - " pl.col('body_mass_g').median().alias('median_mass'),\n", - " pl.col('body_mass_g').quantile(0.9).alias('p90_mass'),\n", - " ])\n", - " .sort('mean_mass', descending=True)\n", - ")\n", - "mass_by_species" - ] - }, - { - "cell_type": "markdown", - "id": "20426f71", - "metadata": {}, - "source": [ - "### Try it yourself (bivariate)\n", - "\n", - "1) Filter penguins to a single island and compute mean body mass. \n", - "2) Make a scatter plot of flipper length vs body mass.\n", - "\n", - "- **Minimum**: island filter + mean body mass.\n", - "- **Stretch**: scatter plot; try `alpha=0.5` and a smaller `s` if it’s dense.\n", - "\n", - "
\n", - "Hint\n", - "Use `.filter(pl.col('island') == 'Biscoe')` and `plt.scatter(...)`.\n", - "
\n" - ] - }, - { - "cell_type": "code", - "execution_count": 59, - "id": "fc115819", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsQAAAHUCAYAAAA9az6FAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAASdAAAEnQB3mYfeAAAqFxJREFUeJztnQd8FGX6x59kN5veQyBEOkgvgiKcBREVz3Jiu7/n4Ym9997Rs54V+3FnRz07d/Y7FawgWEBAQHoACSEJCaRnN/v//N4w6+xmk8xuZndnd3/fj+swM+++884z706eeeYpCW632y2EEEIIIYTEKYmRHgAhhBBCCCGRhAoxIYQQQgiJa6gQE0IIIYSQuIYKMSGEEEIIiWuoEBNCCCGEkLiGCjEhhBBCCIlrqBATQgghhJC4hgoxIYQQQgiJa6gQE0IIIYSQuIYKMSGEEEIIiWuoEBNCCCGEkLiGCjEhpF3mz58vCQkJMnPmTMtLqbS0VE4//XTZa6+9xGazqXFXVVXJ888/r/6NpZ6+ffuqT6xzyCGHqPOPBTAPcS6Yl9FIpK9FtMuPkFBChZgQg+APif4DpaugoEAOPfRQefnllynHCDNjxgx56aWXZNKkSXLzzTfLbbfdJikpKZEeFgmA9h5eCCEk1NhDfgRCYgwoWqC5uVlWr14tc+fOlXnz5sn3338vDz30kMQS48ePl5UrVyrF38o0NTXJ//73PznssMMMP5x8+umnIR8XIYSQ6IAKMSEB4us+AMXq8MMPl0ceeUQuvfTSmHoNn5aWJkOGDJFocJdoaWmRnj17Gv7OgAEDQjomQggh0QNdJgjpIlOmTFFKo9vtlsWLF3u2b9myRS6++GLp37+/JCcnS35+vvzhD3/wauPPt+/NN99Ullkoo3l5eXLKKafI1q1b/R4bfR1xxBGSmZkpWVlZykK6YMECv76CGzduVNvgWmDUv7E9H2KtrdPplLvvvlsGDRqkzrFXr15y3XXXKYutL19++aUce+yxyscXbXv06CETJkyQ22+/XboCHkD69Omj/v3CCy94XFraO0/993wfXvSv7N9//3353e9+J+np6ZKbmysnnXSSrFmzpk0/OA6+s379evWGAHMBrho4zyuuuEJ27drl9/jBzo9XXnlF9t9/f8nIyOjyw9fHH38sRx11lHoDgDHgIeGaa65Rvtftyauurk616d27t/rOwIED5b777lPz3xdsmzVrlgwbNkzJpLi4WJ1zdXV1G/ljTp1xxhnq31jq3ZMwd30J5Hfiyz333KP6ffTRR/3u//XXX5VL1H777efZtnv3bvnrX/8qI0aMUL81/OYgr//7v/9Tb4cCobGxUbn19OvXzyN3/A78/W60h+4jjzxSnSfkuPfee8v111+v5OgPjAftfe8LvuzcuVPJD8f3d/3AMccco2Rl5By1+wLent1xxx2qX4wXv4l//OMfnnZPP/20jBw5UlJTU9XvBG/d8EDrC36HJ554ovqNoC3O5YADDpA5c+b4PT5+g+eee66ak2gPeeE4559/vlRUVHjaQc649mPHjlW/bcgAc/G4446TTz75pNPzJLEJLcSEmID2x0RTKH/44QelqFZWVsrUqVPlhBNOkPLycuVeceCBB8o777yjFBFfnnzySfnPf/6jFCP4wn777bfy2muvydKlS2XJkiXqj6dewcQx8McHfzTwx2fZsmUyefJk5dccDk499VQ1jt///vfqj9UHH3wgf/vb36SsrEyee+45T7uPPvpIjj76aNUG5wbFCLKBOwbOWXNDAVB+oChAyfWnCPly+eWXq3ZQvEaPHi3Tpk1T28eMGRP0eb399tvy4YcfyvHHH6/+yEP2b731lnKN+eabb2Tw4MFtvgPl94svvpA//vGP6g8rlE28NYB8vvrqKy9/5mDnx4MPPqhcQ/BggevcnkJkBCgskDuUBig9hYWF8tNPP8kDDzygriMUKFwvPZhrGDcURlxzu92uxgzlrKGhwes6gosuukieeuopZbmHouJwONT8XrRokeorKSnJ68EiJydH/v3vfyv56a8ftgf7O/HHX/7yF6WQ4gEKb3V8gcIFBQ1BmtrvGwomrv3EiRPl7LPPVue+efNm9ZBy0EEHybhx4wzLHnMEDz54yIIMcM546Pnuu+/UeekfTP/+97/LBRdcoB7MTj75ZHWdcEw8hLz77rvy9ddfe8kHY4QCDKUP8wrKIWSCeex7X4AyiAcJ/FahCOJNl+9DG367OLdAzg994ppgDuP88PCC649/Y45B7phzMCbgfDEXoZTiYVoPzhsPUwcffLAUFRUppRZz87TTTlPuanhA0di2bZt6gMEDKI6LeyLm5IYNG1RsAR7E8NCpzbVXX31VPdxgLkB5xpzG7xTnC/mROMRNCDEEfi7+fjL/+9//3AkJCeqzceNGd3Nzs3vAgAHu5ORk9/z5873abt261d2zZ093jx493A0NDZ7tt912m+o7MzPT/dNPP3l9509/+pPa99prr3m2uVwu98CBA9X2Dz74wKv9U0895RnrvHnzPNs3bNigtp1++ul+z2/SpEltzg/fxzaMz1/bsWPHuisqKjzba2pq1LknJia6t23b5tl+wgknqPZLlixpc9wdO3Z4rWvj7NOnj9soHZ3bc889p/ZhqQf9+x5Da4vPu+++67XvkUceUdsPPfRQr+04Jrbn5+er66+/Rtp533HHHZ7tXZkfaWlp7h9++MEdCP6u62effaa2TZw40b1z506/Mrj88su9tkNW2P773//eXVdX59m+fft2d3Z2tvo0NTV5tn/xxReq/d577+11jMbGRvdBBx3k9xq3d62C/Z10xBFHHKHaL1u2rM2+YcOGuZOSktzl5eVqHcdC22nTprVpi+tcWVkZ0LUYNGiQ13fq6+vdEyZMUPtefPFFz3bMJ4fDoc535cqVXn1dcMEFqv0555zj2dbS0uIePHiw2j537ly/89f3vrB48WK17cQTT2xX3rNnzw7o/Pbdd1+va75u3Tolz5ycHHffvn3dW7Zs8exDO/x2CgoK1G9Dz9q1a9scA/MHv0G73e7Vz6OPPqqOjfP0Bfclbc5WVVWpe/W4cePcTqezTVvtmpP4gy4ThAQILDn43HTTTcrCA8sR9GVYKmHVxKv2devWySWXXKKsV3pgKbv22muVz6u/oC5Yq/CKT88555yjlrCq6a1Aa9euVVZCWOr0wBKDV6rhAFYqWBg1YMX685//rKxrsHb5AkuML74Be7Aew3IcyaA3WNJgwdIDCxOs8J999pls2rSpzXcuu+wyj+sGSExMlPvvv18tn332Wc/2rswPXNt99tmny+enuQrgNbav9RXWM1hn2wtOxHf11xEWS1h0Ya2G1U4DVkCA34n+GLASw2WhKxj9nXSEZv3VxqmBefvzzz+r669ZFDuav7i+sLQGwi233OL1Hbw90GSinyuwVMPSi7nn68t/1113KZcIWD/hgqHdF3ANYFHFNfE3f33Zd9991QdWasw7DZfLJc8884w6xp/+9KeAzu/ee+/1uuZwecCbD7ji4NzxG9dAO7zxwBsSX5cXf+PF/MGbB7hr+fuN+LtGuC9p22F9x/0abxFw7XzxveYkfqDLBCEBovm84saKmzlel5511lkyffp0tV3z1YPS5C9/r+aHCqXP97U4/jD5Ar9czd9P48cff1RL/JHxBTd5+L7+8ssvIb+2RscLJRluCPB9hc8lFHn4AsJ/0Be8Vo10IJ+vogrgUwp5Q5mF/PXKb3vfgSIAecClA8oA5ktX5gd8Zs0AY4Cc33jjDfXxBUrYjh071CtqvYKQnZ2tXsF3dY7CdxwuB6Gedx0BdxicD5ROKHC4vnoFWe+Djtf2eEjAa3ZcNyibOC+MAwpaoPibK7iPQCaa3DTXGuDPBQoKNR6O4KazatUq5S6kte9s/vpy4YUXyplnnqmU8RtvvFFtg2sCXCbgtgB/9a5eHy3g1Z/rhaYg43j631VJSYl66Ibii3/X19d7fU+vQMN9BmOHsgx3Jbgi4R6Da6d3QYEbEBRwuJvgmsK1ArLHvQluGyR+oUJMSIC0F3yioQVv+FM09NTU1LTZ5mutA5riAIuNhuY72r17d799t7fdbIyOF76M7733nvKBxR9d+EVqfxxhGfP1XYw07ckPgYDAn+9uR9+BEoXvQF5dmR/a8bsKxgALW2cBjRiDXiH2d72DmaNQzrpiiTM67zoCFkP48sJK/t///le9aYFfM5Tebt26eb15wXjxZgC+rvCH1XxdYT2FpRlzOBClsSOZwP/eV4bwn/WHtl0LguzsvtDe/IHP71VXXaVkAX9wPFRrv9HzzjtPAgUPGu1dn472Qf76ADk8AOIBBworfNfxXcgJD5h4cNEs4wCKNN4O4CETfsB4ANcelK6++movX3H4m0PRRoCq5vcOKz3e+MGHPlz3T2It6DJBiMloN3y8goTy3N7HNwApELRgp+3bt/vd72+79noQipA//GUWMBME1UGpwB84WHwQhLZixQr1ahqvqK1Ee3LVXin7+6Nu9DtdmR9mVTnDGGBh7Oj4+Phawc2ao1Ba9VH/kcLXbQIPbRgXgkX1AX8A8nr44YdVIB2s+P/85z/Vm4zHH39cWVEDoSOZ6AMZtbmid2XQg0AyfTtt2dlc9PdwAIs4FE08HGjBdLCawvIcCZCxBfKA2waCCOGqgyA6KLyw/vpj6NChStnF9+D6Ass/3LfgzoR+9OeLfvAWDZZnvCWA9RxLKMUkPqFCTIjJ4HUwQHaBUKH5kSIq2hf8AYAvoS+azyL+oPuCyOxwuFho/nx4BYw/eHjFidfzyOhgJT7//HO/Cosmb39+vP6+AysX5I2UTppVMxzzozMwBjyY4IEkEnN04cKFfh/MNLcFo1beroJX6kgZiIcTWFc1xVhTlNsDbiNwk8I1h2UY3w8Ef3MF8wEy0c8t7d/+Si3jARbZI2DZhCIIkEbMyPz1B5R6PHDBMgxlH+2DsQ6bBWIkAFwafPF3fr4WZ7x9giUfFn+AbCj+gAUZLl1ws8BcgIys8LBGwg8VYkJMBv6FCAZ54oknlB9eez6cyOfalT/kOAbSgPkqk7Nnz/ar3OL1LixaSNOkt8jiD9+VV17Zxj/PTGAR9te/ZsnS++7htSl8Iv35OoYLWLJhLdQDSyDGBP9nf5ZTpH3TB9vhwQT5erHU8uuGa350BqzzWiAa0k35Ultbq5TWroB0Vlrwl97FBA9Amp+qL5obBax24QLKL9JzIZUbrseoUaPaPPAgdZe/hwc8VOC1vb9Aro6ApVPv64zj33DDDerf+rmCuARYqh977DGPgqiB4DQ8yKKNlmYOsQNICQi/Yl8lXZu/7QFlEGnQMO+RJxgPcPD3jxRajmrfhwEorlDYfYG7hD/LuO89Br7xSAnnb84j1zSU6WD8wkn0Qx9iQkwGf8Dgv4bXenATwB8pBG/ghgxrIfKPwnKI153BBnHA/QF/FJDhAsEkWh5i5PhEnlr4P0JR9o2ihoIGyxYUauQ0hXUJSjWUULwaRR7XUAD/RLyORS5U/KHDHxwk+ofiCeUSPoz6QBlYvIzmIQ4FCLpB0BU+sAZCLlCWkFEDipM/IFNcZygReHWNP9z4HixVyBwRzvnRGVB88DoZShgUIQTvIfczfIah1MMCh1fIeG0eLAjsQlYMPKANHz5czVGcO4KZIB8EWfnOT+T4xTkjfzNyNGu+nMjI4c9NxQyguN96663KRQW/A3/WYVxHzAVcS+SuxdihWEHpxHd88+d2BuY3ZKLPQwxlFfMBOXY18FuBLBAoBusvfJ7h34zrg4cmPODCF1YDFl64BsAnH/LW8hBj/MgzjPtFR9cUwXVoByUSMo9kkBnGgvzIuE/hXBB4t3z5cjV+yAGuEXrgD4yHTMw7nDPeiEGmmG94YEAWIO3+gjckuAaQKSzEeLDAgwBcSuBrDOMBiUMinfeNkGjPQ9weyM963XXXuYcPH+5OTU11p6enq9zByPf50ksveeXc1PJ96vODGsmxu3DhQvdhhx3mzsjIUJ8pU6a4v/nmG/dFF12kvvPjjz+2+c4///lPlWcV+U27d+/uPvfcc1XuzWDyEPvDXy5Z5IY95ZRT1PlDDsirCrnceOON7rKyMsvlIcYSeYiRGxa5f5FjFzmFV69e3aZ/LQ8xcq0+8MADKg8scgwjn/Bll13mrq6uDun86IyOrtWXX37pPvnkk91FRUUqTyxywY4ePdp9xRVXqPy0ncmrs/EhR+9DDz2kZIL5huNceOGFKhcs5iuO5cuHH36o5A55aL85XN/O5NBZnu2OwO8G30Vu29LS0jb7N2/e7L7hhhvcv/vd79RvBudSXFzsPvLII9vkATdyLZBj+qabblI5edFXv3793DNnzvTKPa3n448/dh9++OEqjy/aI4/1Nddc0yaHtMZ3333nnjp1apv7QmfzCHl5MQfQZvny5YbPy/f8/KH9TrRrqae9cX399dfuyZMnq/PGeRxwwAHud955x+99CffC888/3z1q1Ch3bm6uOyUlRclpxowZXrmmIbPbb79d9YvfKOSJvN8Y+yuvvKJyOZP4JAH/i7RSTggxF1gr8VoQr6rhs0uMgVKxeGUNy1RnpZ810A6+p3it3tVSyvECgtKQKxtvBjQfTxJ58GYC1lXcPyLp405IJKAPMSFRCnxM/WWGgFKHoDqkKaIyTCIJXkHDh9p33mqvr+GGQKwDUo7BRoYiHoTEG/QhJiRKQeARgn/gLwirDiLUkdQfUdIIiEHOX0IiCfxfYQGG7zhy5moV+JDWC37u8A8lkb+PwP8WVnu8GUEsAa8LiUeoEBMSpSDgCOmCEGCDwDhEuyPxPl75o1yuv7KnhIQTPKwhoAu5bREkhwh+uEogcAlWYrPyKpOuuUkguBIBdLheTz31lN+SxoTEOvQhJoQQQgghcQ0fAwkhhBBCSFxDhZgQQgghhMQ1VIgJIYQQQkhcQ4WYEEIIIYTENcwyEUKQIxYZAFAaUqs1TwghhBBCQgeyLm3evFmV8kYaUiNQIQ4hUIanTZsWykMQQgghhBA/zJ07V4477jgxAhXiEALLsHZBUDjBHzU1NbJo0SIZP368ZGRkhHI4hDKPKJzrlHs8wflOmccLNRbUY9auXasMkpoeZgQqxCFEc5OAMjx8+HC/bXbt2qWqNw0dOlSysrJCORxCmUcUznXKPZ7gfKfM44VdFtZjAnFXZVAdIYQQQgiJa6gQE0IIIYSQuIYKMSGEEEIIiWuoEBNCCCGEkLiGCjEhhBBCCIlrqBATQgghhJC4hgoxIYQQQgiJa6gQE0IIIYSQuIYKMSGEEEIIiWuoEBNCCCGEkLiGCjEhhBBC4oa6JqeUVjeoJSEads+/CCGEEEJilGZXi8xbVSbfb9opuxuaJTMlScb1yZXJQwolyUb7YLxDhZgQQgghMQ+U4Q+WbZOaRqfkpDpk3Y4a2b6rQRISEuTwYd0jPTwSYfhIRAghhJCYBu4RsAxDGR5WlCXFualqifXvNlbSfYJQISaEEEJIbLOr3qncJGAZhkUYYIn13Q1OtZ/EN7QQE0IIISSmyUq1K5/hqvomcbvdahuWWM9Msav9JL7hDCCEEEJITJPmsKsAOvgM/7xtl7IMQxnOSLbLvn3z1H4S33AGEEIIISTmQTYJuEnAZxhuEgO6ZShl+JDB3SI9NGIBqBATQgghJOZBajVkkzhgYL7yGYabBC3DRIMKMSGEEELiBijBVISJLwyqI4QQQgghcQ0VYkIIIYQQEtdQISaEEEIIIXENFWJCCCGEEBLXUCEmhBBCCCFxDRViQgghhBAS11AhJoQQQgghcQ0VYkIIIYQQEtdQISaEEEIIIXENFWJCCCGEEBLXUCEmhBBCCCFxDRViQgghhBAS11AhJoQQQgghcQ0VYkIIIYQQEhQNzS6vZbRChZgQQgghhAREs6tF/ruiVOYs3KTWscQ6tkcj9kgPgBBCCCGERBfzVpXJB8u2iTTXy4AskZLKOinZvU0SEhLk8GHdJdqwpIX4q6++kqOOOkpyc3MlNTVVBg0aJH/961+92vzwww9y2GGHSUZGhuTk5MgJJ5wg69ev99vfY489JkOGDJHk5GTp16+f3H777dLc3NymXVlZmcyYMUMKCgokLS1NJk6cKJ9++mnIzpMQQgghJNqoa3LK95t2Sk2jUwZ0y1DbsMT6dxsr1f5ow3IK8SuvvCKTJk2S7OxsefHFF+WDDz6Q6667Ttxut6fNqlWr5JBDDpGmpiZ5/fXX5dlnn5VffvlFDjroINmxY4dXf3fddZdcdtllSmH++OOP5cILL5S7775bLrroIq92jY2NMmXKFKUAz5o1S/79739L9+7d5cgjj5TPP/88bOdPCCGEEGJldtU7ZXdDs+SkOpRFGGCJ9d0NTrU/2rCUy8TWrVvl3HPPlfPOO0+efPJJz/bJkyd7tbv11luVtfe9996TrKwstW3cuHHKkvzAAw/Ifffdp7ZVVFTInXfeKeecc45SggEUaViHb775Zrn88stl2LBhavszzzwjy5cvl2+++UZZhrXjjh49Wq699lr59ttvwyYHQgghhAQf3NWqGQQPLJxQ6rJS7ZLmsJSqZAmyUu2SmZIk63bUiDsrWW2D4bKqvklZirE/2rCUhfif//yn1NbWKotwezidTqUIn3jiiR5lGPTp00cpsO+8845n20cffSQNDQ1yxhlnePWBdVy4uXPnerbhe4MHD/Yow8But8v06dNl0aJFSlknhBBCSOwGd2l9zfpkjTz66S9qGc2BYqEizWGXcX1yJSPZrpRigCXW9+2bF5UPEZZSiL/44gvJy8tTLhFjxoxRCmlhYaGcf/75smvXLtVm3bp1Ul9fL6NGjWrzfWxbu3atUoIBLL5g5MiRXu2KioqUn7C2X2vbXp9gxYoVJp8tIYQQQswK7kJQF8AS6/NX7wi6Lyh3TU63WgbbV6wzeUihHD2qp/TOS1PrWGL9kMHdJBqxlAoPK2xdXZ2cfPLJcsMNN8gjjzwiixcvlttuu00prF9++aVygwBQnH3BNlh+d+7cqZRetIVrRXp6ut+2Wl8A/26vT21/RyAgz9d/Gco5qKmp8Sj0vsAirl+S0EOZRwbKnXKPJzjfwwPcI5ZvLFWZDobkO0SaRS1XVdTLsg3bZHR3h6Qk2QLua3T3DOUTC3cAKMWB9hUv7L9XqgzKKJClP26SE0YWSF5OqtTX1kh9hMcFvSuqFeKWlhZl3YUCfP3113t8fh0Oh/L3RcAbsj8AzYnbH/p9RtsF2tYX+Dwje4U/4HJRWlra4ffRhoQXyjwyUO6UezzB+R56BuADD8o9yaOKmrdKEdadIgu+2hRcX42/Gbh6B9lXvLH0x+/FKpSUlES3Qpyfny9r1qyRqVOnem3//e9/rxRipFo77rjj2rXYVlZWtkY55uR4+oOCDauzpkjr2yIQT3/s9voE/qzHepC9ApZtXwvxtGnTZPz48TJ06NB2rQi4YaKNP0s2MR/KPDJQ7pR7PMH5Hh5g1YXPMNwkYBmGMrwtqVhWVTSpV/jTJ/QJyEKs9YXAMGUhdre6TfjrC+1rGl2SkWyLWctxg4FztOJcX7lyZXQrxPDXXbhwYZvtWsq1xMREGTBggMpNvGzZsjbtsG3gwIGSkpLi5TuM7fvvv7+nHay15eXlMmLECM82tG2vT6Bv6w/4OuPjD+RK1gcA+gOTqLM2xFwo88hAuVPu8QTne2jBX80RfRtVQQi4ScAyDGVYklJlZL8iKczPDaqvpdsbVQoxZE3ISPbuCwF28DVGHl6kHkO2BQSYwac2yWap0KygaQ7iHK0016F3BYqlrhwyR4APP/zQaztyEYMJEyaoQLtjjz1W3n77bdm9e7eXeXzevHkq37AGcghDOX7++ee9+sM6nvxgvdU4/vjjVTCfPr0aMlrMmTNHKdM9e/YMwRkTQgghxCrBXVpfsBA77Ilq6dtXPATezYuDc7S0hfiII45Qyu4dd9yh/ImhAH/33XfKN/eYY46RAw88ULXD+n777ae2wdcYbhHITYzMEVdddZWnP7g5IN/wLbfcov6N/hGkN3PmTDn77LM9OYjBmWeeKU888YRye7j33nuVtRd+watXr5ZPPvkkIvIghBBCSMfAYolSwQh6g58vXBsCsQz76+uAgfl+8xDrK7QNK8pSxrWeOSny87ZdqkIbvheNKcf0xMM5Wt5CDF577TXlLzx79mzlO/zUU0/JFVdcIW+++aanDcowz58/X5KSkuSkk05S5ZbhKoG0bd26eT8R3nTTTSpbBb4PhRhlnKFEQ/nVg2wUCNpDLuNLLrlEKebbtm1T1mpUziOEEEKIddF8XM3w54XC1yM7pY3iF4sV2nzZFQfn6A/LqfjwD4aFFp+OQECcUcvtpZdeqj6dgVLNL7zwguGxEkIIISQ+K7TBaqoF3kVzhbZ4PEd/xOZZEUIIIYSEqELb9l0NyoXgt8C76K3QFo/n6I/YPCtCCCGEkBCAwDtYTeFPCxcCWE2hKEZrhbZ4PUdfqBATQgghhJgUeBcLJMXBOfoS22dHCCGExAHIDGA1xcWKYzITnFMsnleg54jiHdrSGlmIgyO2ryQhhBASw1ixSIQVx0RCd52XbyxVJa9R5Q+FTaL1OlMhJoQQQqK8gAJyxiL4CZkBEAwF/0+88uaYSKjnnjTXy4AsUSWvUeUvknOvK0SfCk8IIYSQNgUUinNT1RLrCIbC/nBjxTGR0F7nAd1ayyRjGc3XmQoxIYQQEoVYsYCCFcdEzGdXDF5nKsSEEEJIFBdQQI5YFE4AWgGFzBR7RAooWHFMJHjqmpxSWt3QxuKrv87NztagOiyj+TpH34gJIYQQYskCClYcEzE/MDLNYZcxvXJkyeYq+XJNlfTvJ/Llmh2SlJoh+/TOicrrHH0jJoQQQohlCyhYcUzE/GBNt6d1gtfyt+3RBRViQgghJEqxYgEFK46JBB8YmZCQID1zUpTFHw85uK5g6eYq5R4xfK8CEWetHDSoQFaUN8uSkiqZtHe3qLvm0TVaQgghhERFkQgrjskosV5UxIyAud172iTZE0WcIkl2m+SkJnjaRJvcomu0hBBCCCEhgkVFxBMwBzcJWIahDGuBkXB/0QLmtDburGS17q9NNBF9IyaEEEIIiZNCJ1YNjBy3pw1k1DtL1DIjOTVqgyejb8SEEEIIIRHwnY1GRS9UgZGT97RZtmGbcpnonZcmI/sVRW3wZHxcWUIIIYSQLvrOxotCnGQgMFJrM7q7QxZ8tUmmT+gjhfm5Eq2wMAchhBBCIkZ5TYP8tKVKLSNJJIuKtFcAI9KkOezSIzulwweBhmaX1zJaiY9HHUIIIYRYCih/j3+2Vr5eWy61jU5JT7bLAQML5OJDB8ZNUZFoDuKr23P9VmwslROKRG6Zu1yG9+0RsevXVaJvxIQQQgiJeqBM/WfJVmlwtki6wyZbq+rVOtwUrpk6OC6KikRzEN/je65fps0pUiRStrtR1kX4+nUFKsSEEEIICStwj4BlGMpwv/w0SUxMlJaWFtlQUSdfrdkhZxzQRwoyUmK6qEg0B/GV667fkHxcpwYpzkmRJdubInr9uoK17fGEEEIIiTl+rWpodZNw2JQyDLDEOrZjv9V9Z8NVAMOK/Grx6xcMVIgJIYQQEtbgLlhC4TNc2+RSlmGAJdaxHftjXaaRDOIL9/Wrs2jQoB7rSpsQQgghYSHcwV14nY4AOvigwk1CWRabXJJiT5QDB3WLutftwcg0EkF8obh+W2ENLhS1TLHbva5fNAUNWlfahBBCCAkLkQjuQjYC9A+fU7xmL85xKGXqoskDJF5kGu4gvlBcv+UozCENUpiZLIf2K/K6ftEUNEiFmBBCCIljIhXchT6RjQABWPA5xTFjwTIciEzDGcQXqutXUpovP377tfx12gjp3aMgaoMGrTMSQgghhMRdhTYowbGiCAcrU/w7nMohskSY9RCSk+bwWlplXgWKdUZCCCGEkLCjBXfhdTYUJCgtWnAXXuFbObjLqlhVpuEshpJlURm0h7VGQwghhJCwEs3BXVbFqjINZzGUNIvKoD2sNRpCCCGEhJ1oDu6yKlaTaSSKoUy2mAw6ggoxIYQQEudEc3CXVbGaTI0U0ygwWSG2mgw6wlpJ4AghhBBi+QptViy00NDs8lp2BSPnZ0UZRKIYSoMBuYej8l9Xse7ICCGEEGIprFhoQRvT8o2lggy4cxZukhF9G4Mak5HzMyoDq8nK7GIozSbK3QpQISaEEEKIIaxYaEEbkzTXy4AskZLKOinZvS2oMRk5P6MysKKszCyGMs9EuVsBS6nw8+fPV4L091m4cKGn3YwZM/y2GTJkiN9+H3vsMbUvOTlZ+vXrJ7fffrs0Nze3aVdWVqb6LigokLS0NJk4caJ8+umnIT1nQgghJBrwLbRQnJuqllhH0FQkXAf0Y0LAFsAymDEZOT+jMrCirPTFNJ6Zsa889H9j1BLraQG6Mpgpd6tgSQvx3XffLZMnT/baNmLECK/11NRU+eyzz9ps8+Wuu+6SW265Ra6//no54ogjZPHixXLzzTfL1q1bZfbs2Z52jY2NMmXKFKmqqpJZs2ZJYWGhPPHEE3LkkUfKJ598IpMmTTL9PAkhhJBowYqFFswck5G+gJHjWVFWZhZD2WXx8wsGS4520KBBMmHChA7bIDKyszYVFRVy5513yjnnnKOUbHDIIYco6zCU4ssvv1yGDRumtj/zzDOyfPly+eabb5RlGEApHz16tFx77bXy7bffmnZ+hBBCohNYvqweLR+q84tkoYX2xqUfkzsrWW0LdkxGz89IG6sXpejqPM7SnV9zepLa1ux0SVV9syXOLxiib8QB8NFHH0lDQ4OcccYZXtuxftNNN8ncuXM9CvE777wjgwcP9ijDwG63y/Tp0+XGG29UFuXi4uKwnwMhhJDIY7UAqUicXyQKLXQ2Lv2YoJz1zhK1zEhODXhMRs/PSBurFqUwax6nOewyuleOLNm8U75cUy39+4l8uaZcklLTZEzvnKh8WLTkiC+66CI55ZRTPH68cHk48MADvdrU19dLjx49ZMeOHVJUVCTTpk2TO+64Q/Ly8jxtYPEFI0eO9Pou2sNPWNuvtT3ooIPajGXUqFFquWLFCirEhBASp1gxQCoS5xfuQgtGxqWNadmGbSJOkd55aTKyX1FQYzJyfkZlYMWiFGbO4wTP/917tmCZsGd79GEphTg7O1suu+wy5daQn58va9eulfvvv1+tv//++zJ16lTVDm4M+Gh+xZ9//rk8/PDDKgAOPsIZGRkelwkE0qWnp7c5FhRn7NfAv/XKtL6dtr8jEJAH5VwPxg9qampk165dfr9XW1vrtSShhzKPDJQ75R6tIL8qUkshmn5094zW199ZyUqZgBI2urtDUpJsUTvfAz2//fdKldHde0hNo0sykm1qX31tjdRHcFwY06CMAln64yY5YWSB5OWkBj0mI+dnVAbhklWo5nFHfa0s2S5FaW4ZWJwt4qqTQwdmy9qdTvl503YZW5RiuK9QAL0rUBLccGqxMAhyg4UXiunSpUvbbffWW2/JSSedJA899JBcccUVatu5554rL730krIm+wL3CGScgFsFcDgcctZZZ8lTTz3l1W7BggXyu9/9Tl599VVltW6PmTNnquwV/nj00Ueld+/ehs+ZEEIIIYQER0lJiVx66aXq7f/w4cOjz0Lsj5ycHDnmmGPk6aefVoqtv0wS4Pjjj1eWYH16NliZ4UNcV1en3C/0VFZWyrhx47za+rMCox3wZz3Wc+GFF8rJJ5/cxkIMV47x48fL0KFD/X4P1oNFixapNv4s2cR8KPPIQLlT7tEKrGEoOoA8q3jtrQVIKZ/VvDSZPqGPXwtxtNzbgzk/K46rM5k7W9yyaH2F8ulVJYyT7SoV2vj++WJPjNYX/ZG5zg26vobkO6SoeatsSyqWVRVNEZ0zGitXrgz4O5ZXiIFmxNZSe3TUTqvPrfcdXrZsmey///6e7aWlpVJeXu6Vyg1t0c4XbZtv2jdfkKYNH3/AhSMrK6vD7+PH21kbYi6UeWSg3Cn3aAN3ZlTgQtGBpdsbdQFSqcpXtTA/N6rne1fOz4rjak/m/11RKh/9Uu3xn920u0lKdleLLSU9JvzAw3mds3R9raqol6IsUcqwJEV2zmhorrMxpRDv3LlT3nvvPRkzZoykpLSfM+/NN99UlmB9KjbkEMZ3nn/+eS+FGOtQrmG91VuYYeVFejWtrdPplDlz5qj1nj17huwcCSGEWBsrBkjFw/mZNS7fQhnoE+nQYC1G3wcMzI/KzAiRvM6TTQxmtAKWuvqnnnqq8rXdd999VRaINWvWyIMPPijbt29XSizYtGmTagd/3oEDW0sQIqjukUceUX4iZ599tqc/uDkg3zCyVODfWmEO+PuinZZyDZx55pmqEAfcHu69915l7X3yySdl9erVqjAHIYSQ+AUpqWBFhOIUi3mIrXp+Zo0rFgtJRPo6J+3pC8F4C77apNwkIm0Z7gqWuvpIcfbaa68pf2FECEKJRbo1BMbtt99+qg1eg3Tv3l0Fz0FRdrlc0qdPH+U8jXzBvn5DyDecmZmplN0HHnhApWpD1Tps14NsFMhSgSIcl1xyibI2wyr94YcfskodIYQQBZQHKypOsV4wxIjc4deqLX0dJmK9UEYkSdnjKxxJn2EzsJTUoaji0xG5ubny9ttvB9QvlGV8OgOK9gsvvBBQ34QQQkikMKvQQjQXHtHGjpRiA0RUsBf8WyNdVMSKco/m6xxXCjEhhBBCwl9oIZoLj2hjR37dAVmiMh8g2CvSRUWsKPdovs6hhgoxIYQQEoWYFSgWzQFn+rGj2IQ07lCKLrIo+I7dan7S4ZZ7NF/ncBDf9nFCCCEkSjESKBbOfiJBMGOH0tcjOyXiyl+45R7N1zkcUCEmhBBC9ljQSqsb1DIa0ALF4Aur5evXAsUyU+yGA8XM6icShGrsZs6F9voKt9yj+TqHg/g+e0IIIXFPtAYamRUoZtWAs0DHriquZYlaothEMGM3cy501le45R7N1zkcxPfZE0IIiXuiOdDIrEAxKwacRaJAhJlzwUhf4ZZ7NF/nUEOFmBBCSNwS7YFGZgWKWS3gLBIFIsycC0b7Crfco/k6hxpKgRBCSNwSKxXMrFowxIpFKdo7nplzIdC+wn39onm+hApKgxBCSNxi9QpmsexHbXZRkY4Kcxg5nplzgfMq+oiPXzohhBDiBwYaRc6P2uyiIp0V5ujseGbOBc6r6MO64bOEEEJIGICF8OhRPZUV0GFPVEusx0ugka+/a3FuqlpiHf6uoUhDZ9Yx9f3gugEsffsxejwz50K8z6togxZiQgghcU28BxpFwo/arGMa7cdoOzPnQrzPq2iDFmJCCCHEQhXMwk0kCjYEesyuFreIZFEKI/Mq2orCxCLx9asnhBBCSMT9XY0eM5DiFh0V5jDreGYTrUVhYhEqxIQQQkicE4mCDUaOGUhxi84Kc5h1PDOJ5qIwsQYVYkIIISTOiYS/a2fHDLS4RWeFOcw6nllEe1GYWIOSJoQQQkjECja0d8xAA+9Skmxey1Afr6vESlGYWIEOKoQQQohFieZgq66OXR8I1+R0KUsqlqEKhAt34F0wx4vm+WB1+OhBCCGEWIxoDrYya+ywjo7plSNLNlcpP9vEhARpcbulR3aq7NM7x3TrabiDCwM5XjTPh2iBCjEhhBBiMaI52MrMsbfaTUGC1/K37dEdXGj0eNE8H6IFKsSEEEKIhYjmYCszx46+lm6uUu4DY3v3kCanWxz2BFlTViNLSqpk0t7dTJdDuIMLjRwvmudDNEEJEkIIIRYiksFWUL66ogiaOXZ9Xw67TbSvtddXQ7PLs8yS6Aou7Oh4DL4LD3Q8IYQQQixEJKqqwUf1vytKZdYna+TRT39RS6xje6TGbrQvbexzFm5S61gGM3arEskqe/EEpUgIIYTEeeU4s3xUzRy70b60sUtzvQzIEimprJOS3dtixr82EvMhHqEUCSGEEIsRzuAus31UzRx7Z33pxz66e4ZI4w7VZun2xpjyr41EJcF4I/pnCSGEEBJjhDO4y2wfVTPH3llf8eJfG4lKgvEGfYgJIYQQiwKlp0d2SqfKT1cKNrQpgNFgTgEMo2PvSl/xVtzCTJkSbyhRQgghJEoxo2ADlKvRqgDGTvlgWakkJoi0uEV6ZCfLmBAUwAiVfy38nntniVpmJKeyuAUJiC7P8l9++UW2bt0q9fX1UlBQIIMHD5bs7OyudksIIYSQMAXD7XE22FPy4relVg4jGvxrl23YJuIU6Z2XJiP7FbG4BQm9Qrxw4UJ5+umn5cMPP5Ty8nLPKwpMyMTERBkzZoxMnz5dZsyYQeWYEEIICQFmBcOhnyWeAhhF0uRyi8PWWgDjx5IqOTgEBTBC4V87urtDFny1SaZP6COF+blebVjcgnRGQDN8yZIlcvnll8sXX3whQ4cOlZNOOknGjh0rhYWFkpKSIpWVlbJ+/XqlMN90001yyy23yI033ihXXnmlOByOQA5FCCGEkA4wK6As0AIYViUlyea11BMvwXckeAK6+vvvv7/8+c9/loceekgpwh1RW1sr//rXv+Rvf/ubOJ1Oufnmm7swTEIIISQy1desihZQBjcJWIah4GkBZUjLZTQYTt9Pt0yHx0IcaD+RpqNKdWbJisQuAc2A5cuXy6BBgwy1TU9Pl7POOku5TWzevDnY8RFCCCERCziLh4INaDdGBdVVKX/kxIQEaXG7pUd2quxj8aA6/XVevrFUBuypVDeib6PXdWZxC9IZAc1yo8qwHpvNJn379g34e4QQQogVAs7ioWBDa8IykOC1/G27dTFaqY7FLUhHWOoRef78+WoC+/vAL1nPDz/8IIcddphkZGRITk6OnHDCCcp/2R+PPfaYDBkyRJKTk6Vfv35y++23S3Nzc5t2ZWVlyqKNbBlpaWkyceJE+fTTT0N2voQQQkKDbxBVcW6qWmIdymM05qDtKKDsssMGyaVTBqkl1gOxgEMWS/cE1R01soccOqS7WmJ9SUmVpWWlv854GABY+rvOZsiKxC5Bvwc59NBD292HTBNQUvfbbz8588wzpVu3wJ5U7777bpk8ebLXthEjRnj+vWrVKjnkkENUNovXX39dGhoa5NZbb5WDDjpIBf7pj3fXXXep4L7rr79ejjjiCFm8eLHyZ0aquNmzZ3vaNTY2ypQpU6SqqkpmzZqlAgWfeOIJOfLII+WTTz6RSZMmBXQOhBBCIke8BVHhXII9n2gOqgvmOndFViR2CXpGtLS0yJo1a2Tbtm3KJaJ79+5SWloqmzZtkp49eyql9L333pOHH35Yvvzyy4DcLdB2woQJ7e6H8gtrL/rPymp1nR83bpz63gMPPCD33Xef2lZRUSF33nmnnHPOOUrJBlCkYR2GUoyMGcOGDVPbn3nmGeUj/c033yjLMIBSPnr0aLn22mvl22+/DVZUhBBCwgyDqIKTVU6aXWoaXZKRbOsw4MwqgYr6sTenJ6ltzarKXrPlx271McUbQb8ngEKZlJQkCxYsUK4KWG7YsEEplPAbhgK6cuVK5dJgZoYJZKyAInziiSd6lGHQp08fpcC+8847nm0fffSRsh6fccYZXn1gHdGlc+fO9WzD91BURFOGgd1uV/mUFy1apCzKhBBCogMtiAoBZgg427qzXi0DDTiLByCLkcXZsqmiTl7/bou899Ovaon1UXtle8kKAWz/XVEqsz5ZI49++otaYh3bIzV2VNmDlfjLNa11EbDEum+VPauN3apjileCVoiRXxiWWqRi0wPLLrZjP/x1r7nmGpk3b15AfV900UVKGYXCO3XqVPnqq688+9atW6eq4o0aNarN97Bt7dq1SgkGsPiCkSNHerUrKipSfsLafq1te32CFStWBHQOhBBCIguCqI4e1VNZCh32RLXEeqABZ/HAsq3VUlnbJE3NLmlpcasl1rHdXwAbLLJNTrdaYn3+6h0RG7t3lT1pt8qeFcduxTHFK0E/Ii9btkx69erldx+2w88XIJitutr7B9UeKPl82WWXKbeG/Px8pdzef//9av39999XyjHcIEBeXl6b72MbLL87d+5USi/awrUCKeD8tdX6Avh3e31q+zsCAXk7dnhPYIwf1NTUyK5du9rN16xfktBDmUcGyp1yjwT775Uqo7v38LgBoGhDfW2N1If4uNE036vqmmRlSankJDmlR6pNXC1usSUmSoPTKT9vKpWS0m6Sk+ZQ+X2R2gzZHEZ3z2jN5ZuVrJQ4lE1GpTh/RTFCCca0smS7FKW5ZWBxtoirTg4dmC1rd2Ls22VsUYoak1XHbrUxxcpch94VNoUYPsNwMzj88MPb7MN2BKUBKIIIsDPCPvvsoz4aCJI7/vjjlYUXfrxQiDU053l/6PcZbRdoW1+efPJJlb3CH3C5gH91R6ANCS+UeWSg3Cn3eCJa5vuJRXh96m9Pk/z47deeNeT5RWozafzNANQb605RZZMjgWdMrXU5pJdrm/TyMyZLj91CY4qFuV5SUhI+hfjcc89VvsGw/p588slKQd6+fbu89tprKvMDgtkAfIr9uSIYBcr0McccI08//bRylYDluD2LLUpHq+jSPQo42sJ9oq6uTqVR822LQDwNtG2vT+DPeqznwgsvVHLwtRBPmzZNxo8fr0pd+wNPVJhEaOPPkk3MhzKPDJQ75R5PhGu+O1vcsmh9hfKPrm10SnqyXaWXG98/X+yJHRtyNEqr6+Wyfy2R3Y1OyU21e6q47ax3SmayXWadMkYV6YBFE0UvkOcX7idaO1g0e+elyfQJfQKyaJoxdv2YhuQ7pKh5q2xLKpZVFU1eYzJ77GZgxTHFyr0dMWxhU4jhIwyTNLJIoEQzwIWEiwJSnN1www1q25/+9Cc5++yzgz2Mp1+AyTJgwABJTU1VLhu+YNvAgQMlJSXFy3cY2/W+zrDWlpeXe6VyQ9v2+gT6tv6ARVyzivuCwEJ9AKA/MIk6a0PMhTKPDJQ75R5PhHq+IwDro1+qPcVHNu1ukpLd1WJLSTdcfKTO7ZAER6pU7KqVisYWcdgSpUkFdSVKdlaaZGVlS1ZWiiqHjApwKHqxdHujrjJeqozsVySF+blhH7t+TKsq6qUoS5QyLEneYzJ77GZgxTHFyr0delegdCkbNTJJIO0a/HtffPFF+eCDD+TXX39VuX81YB2GEhss8AdGVgnkHIaii2C7Y489Vt5++23ZvXu3l3kcwXso0KGBHML4zvPPP+/VJ9ahXMN6qwHXDPg969OrIaPFnDlzlDKNVHKEEEJItBYfwXppdUOb7Uj1NaF/vnTPSlEW4cTEBLXE+v798rxSl2mBirBgulpa1DKYQEUzC6foxwTaG5MVgyytOKZ4pct5Z+CeAMXTDE499VTp3bu37LvvvioLBPIcP/jgg8oVQ6/UwlcXRT/gSgFrtFaYA9+56qqrPO3g5gC3DhTmwL+1whwzZ85UVmstBzFAAREU4oDbw7333qusvfALXr16tSrMQQghhERjUQqk8EI2AyigaI+8vUhJB2UMVdrQBgoxskrgY7cliNPllrx0h0wcUNAmRZ321ta9J6uDth6KsRtBq0CHIDT43cLVwJ91VWt3wMB8y+T8teKY4pWApA7rqW+atc6A3y/yFA8fPrzTtrAmwwcZ/sJwx4ASe+CBB8pLL72kFGANZK5AmefrrrtOTjrpJGU1RuU8FOXwrYp30003SWZmplJ2sb9Hjx5KicZ2PXD1QJlmBO9dcsklyu8YVukPP/yQVeoIIYREbfERLbWX5pqA9tt3Naj2mmsClGOswzoLhRRlm5Gv2ddS6dsX/F+hROv70iivaZBfqxrU2AoyUoIaeyBo/rad+d1asVKdFccUbwQkfZQvRqYHKIyHHXZYh21h1YUbxUMPPaQqwhlRiKGo4mMEBMQZtdxeeuml6tMZCAx84YUXDPVJCCGEWKH4CJRbBKb95oP6W/ERX9cEKJ5QQNEeyi8sk2hnxFJptC+0e/yztfL12nJPsNwBAwvk4kMHevo0MnZCwklAMw7FKa6++mrleoA8vwcffLCMHTtWuRfAVxcZGVA4Y+HChSriMDc3V7k3ICMFIYQQQswt6etr2YV1VW/ZDdQ1oSNLpdG+oAz/Z8lWqW92KWvtzromtY6210wdbHjskYSllOOPgBRiBMchxzDSWTz11FMqiA4uDnqQAQLlj//+978rn2C4IhBCCCHEOJ35/Wp0Ztk10zXBSF9wk/hqzQ6pbnBKusMmTc4WNUasf/lLmZxxQB+P+4QV/WeNyp3EHkHNPOTUffTRR9UH1dmQWQK+wghq69OnjyQlJZk/UkIIISROMOL3q6c9y66ZrglG+lpbViNluxtVBoqGZhG7LVHl28X6jt2NyqfY15/YSv6zgcqdxA5dnoEIYvMNZCOEEELiGSiB2jLQzKxGfXWNorkmLFhXLjtrWwtWIHuEP9eErrpo5KYlqdLPzS63ZKckSmJioiQlitQ2uVQhDuy3qguD2XIn0QWvLCGEEGLyK/flG0tVWV5UIkPxhUBeuZuZksxoqjSzXDSSbDbplpmslMqqeqenyAcqzxVmpaj9VnVhCIXcSfTAK0sIIYSY/MpdmutlQJaotGSoRBbIK3ezU5IZSZVmlouGVuSjrskldY1OaW5xS6bdLmnJ9jZFPqzmwhCKVHAkeqCHOCGEEGLyK3coUADLQKuvab668M3F6/qtO+vVMhi/XyMV4cysGqcV+RjTK0eGFGXJPr1bl1j3V+Sjq2M3EzPlTqIPXl1CCCHEBF9W/St3+NECLNt75d5RX2alJDPiBgDMdBUwWuTDjLGbraRaORUcCS1UiAkhhBATfFmh2KIIxYpfK6Vyp1P69RD5afNOqWi2y3idu4CRvsxKSWbUDcBMV4Fwj91MrJgKjoQHU6/y5s2bVfEOlFnOz883s2tCCCEkYhjxZdUqvlXUNEqTzSnSQ2TbrkbZ7XKJw27zKFaB+MV2NSWZ0bRroagaF66xhwIrpYIj4SHoq33zzTdLbW2tPPzww2odZZSPPfZYaWpqkpycHPniiy8MlWsmhJB4ghWwoo9AShajEEV+RrLkK2WqQYqyU8TRZJPGZpfH5zXcqb2MuAFY1VXAquMisUfQv7q33npLrrzySi8FedSoUXLTTTfJzJkz5c4775RXX33VrHESQkhUwwpY0YtRX1YsofQOKsyUvRBT11wlo/bKkS01UKpdIfHXNcsNwKquAlYdF4k9gp5VW7dulYEDB6p/V1RUyOLFi1Up56lTp0pDQ4NcddVVZo6TEEKiGlbAil6M+rLq2/XJSlbbbIkJUlXfGLS/Lkoho7ob2vpWeIukG4CZ4zICXRhIqAn6l4EfcEtLi/r3119/LTabTQ4++GC1XlRUJOXl5eaNkhBCohhWwIpujPqy6ttB4e2dJWqZkZwasL8u5szjn62Vr9eWS22jUwXrHTCwQC4+dGBILKRG32CEe1yEhIugZ++AAQPkvffekylTpsi//vUvGT9+vKSmpqp927Ztk9zcXDPHSQghUUs8VcCKVR9po76sWrtlG7aJOEWVSR7Zryhgf10onf9ZslXqm1skNckmO+vq1Tq+d83UwRF7g/HbuFySosbVFNJxERIugr5bnXfeeXLRRRfJiy++KFVVVfLss8969sFiPGzYMLPGSAghUU08VMCKdR9po76sWrvR3R2y4KtNMn1CHynMzw2oL7gjfLWmXHY1OCU1KVEanS6xJ4pa//KXHXLGAX1MdVMw+gajdVw7pLrBKekOmwogxLlg/ctfykwfFyHhJOi78AUXXKCswN98842yDk+fPt2zr76+XmbMmGHWGAkhJKqJZPqocBEvPtJGfVlhPdUvA+kLvrlluxvE2eKWRmeL2BMTpdHVotaxHfvNVDyNvsFoHVejuFpapKFZxG5LlIZml1rfsbvR9HEREk66dBc+5ZRT1MeX2bNnd6VbQgiJOVeBWE4fFU8+0kZdQqAoasusAI+Rm5akKtzB6p6dkiSJiYmS1CJS29SstmN/JN5g/DYut2QmJ6p2toREqW1yKWXd7HEFKveuyJyQoO9QyCSBnMNZWb9Nu9dff11++OEHOfzww5VvMSGExDpGXQViOX1UPPhIG73OWrvlG0tlgIjMWbhJRvRtDMh1JMlmk26ZySporareJQ5bizS53JKUmCCFmclqfyTeYOC4BRnJUl3fLBW1TZIgCeIWtyQmJEhBCMZlVO5myJyQoO9Qp512mqSnp8vzzz+v1h999FG5/PLL1b/vv/9+effdd+Woo46ihAkhUU9H1qlAXQViMX1UPPhIG73OWjtprpcBWSIllXVSsntbQK4jkNeEfvkqdzE+UPgyk1Dpzib798sPiTyNvMHAcXvmpqpzanK6RBJEEtytD3vFOakhGZcRuZshc0KCnr2LFi2S++67z7MOhRh+xI8//ricddZZ8sADD1AhJoRENZ1Zp+LJVSCefaQDqVSntRvdPUOkcYdSLJdubwxoPqDNhAH5UlnXJJW1TZKUmCjNLS2Sl+6QiQMLQiJPo28w7AkJkpliF4fNIYmJCdLS4pYmV4vKtxwJuQMzZE5I0LNkx44dUlxcrP69YcMGWb9+vapMBxcKKMR/+ctfKF1CAiBW01VFM51Zp+LBVcAosewjHUilOrPmg688oYR2JE8j94/NlbWytqxGBhZmSK+8dL9tOnqDgf7hJ9y/IENlvkDAX7I9UZLtNslNdZg+343IE/A3SMwg6JmblpYm1dXV6t9ffvmlZGRkyL777qvWU1JSpKamxpQBEhLrxHq6qmjFiHUqHlwFjBLLPtLBVKpz76lUF+x8MCpPI/eP6vomuf6tZWreQpGFAgvl+t4TR0p2qiMgOWSnOSSttklG98qWJqdbHPYEWVNWI9lpSabPd6NyN0vmJL4JeqaMHDlSnnjiCenTp488+eSTMnnyZM8TXElJifTo0cPMcRISs8RLuqpow4h1qkd2Sky7CgRDLPpIm12pLtBjd/Q9I/cPKMPzV5epTBAOW6JU1Ter9RveXi5P/nlsUHKAEhzq+W5U7mbLnMQnQc+UW265RY455hgZM2aMOBwO+eSTTzz73n//fRk71viPjJB4hT6o1sWodSqWXQWCIVZdf8ysVGeWPI3cPypqWn1poQwXpLemcGtpaZHy2mZZvKFCuVG05z7RFTmYhZHjmS1zEp8Efbc69NBDZeXKlfL9998rpbh///5e+7CNENIx9EG1LsZTUcWuq0AgxLrrj5mV6sySp5H7B3yG4SYByzCUYYAl1rEd+wNRiMM9340czyyZk/imS7MY7hL4+CvrTAjpnHjyQTWaNN9KFsZArGGx6CoQCHT9Cb88jdw/EEAHn2G4ScAyrFmIkRkix5Gk9geDmfPdyG/eyPGMVAckpD1Mmc3IOIFyzb707t3bjO4JiVliPV1VIEnzrWhhpPXXGPHg+hPOwhxG5Wnk/pGW1/pv+AzDTQKWYSjD9sQE2a9ffkDWYbOx4m+exC9dukPdeeedKv9wRUWF3/0uV6tFiBDSPrHug2o0ab6VLYzxbv2NF9cfMwqwmFEkIhB5Grl/IJsEAujgMww3CViGoQzfc8IIiSRW/s2T+CPoO9Szzz4r9957r1x//fVy6623yk033aRe1bz00kuSmpoq1113nbkjJSRGiWUrpNFCBfFgYYxlot31x6wCLGYV5ghEnkbuH0ithmwSRvIQhwv+5onVCPqdBFKu3XjjjXLDDTeo9eOPP15ZjFetWiWZmZlSXl5u5jgJiXnwRwxpvGJJ8TOaWN9oO2JNtFf3eFUPJXHrznq1jBbXH81SCQUUuXWxxPr81TsiMo+DkaeR+weU4MlDukdcGQb8zZOYUYjXrl0rEyZM8EStNjU1qSWsw1dddZXMnj3bvFESQqISzdIFyxYsXECzdKHylm9Bg87aEesCa+rRo3oqC6bDnqiWWO+K6w+siKXVDWoZLktlcW6qWmIdVl3sD2YeNztbXQaxDGYeh0KeVoK/eWI1gv4rY7fbPU+/KNe8ZcsWz76CggLZunWrOSMkhEQtRgsVxENwYaxjputPOIOtzCzAguWYXjmyZHOVfLmmSvr3E/lyzQ5JSs2QfXrnBCSPWHalAvzNE6sR9J1l0KBBsnnzZvXv/fbbT/7xj39Ic3OzCqSDdbhv376mDPCf//ynujmhNLSeGTNmqO2+nyFDhvjt57HHHlP7kpOTpV+/fnL77ber8fpSVlam+oZSj/LUEydOlE8//dSUcyEkHtEsXUiWD7D0Z+mKdYtYvGCG609nLgyRsFQanZ+tPYAEr+Vv2wMjFl2pNPibJ1Yi6F/YUUcdJV988YWcfvrpyo946tSpkpOToyzHNTU1Kuiuq8DKfPXVV0vPnj2lurq6zX64Z3z22Wdttvly1113qcp6CAA84ogjZPHixXLzzTer/vWuHY2NjTJlyhSpqqqSWbNmSWFhofKVPvLII1UlvkmTJnX5nAiJN4wmzY91ixixZrCVmQVYMPalm6uUIj18rwIRZ60cNKhAVpQ3y5KSKpm0dzfOaR38zRMrEfRdBZkl9JXpvvnmG/nXv/6lbl5HH320TJ48ucuDO//88+Xggw+WvLw8efPNN9vsh/8y/Jg7AinhEOx3zjnnyN133622HXLIIco6DKX48ssvl2HDhqntzzzzjCxfvlydCyzDAOcxevRoufbaa+Xbb7/t8jkREq8YTZof6ynOrFR4xIroXRhQbrix2SXJSYkhTeFmVgEW/diT7ImqjHCS3SY5qQl+x8650LlMCQkXps1AuE3gYxZz5syRzz//XH7++WeluAbLRx99JA0NDXLGGWd4bcc6UsXNnTvXoxC/8847MnjwYI8yDGDxnj59usqoAYtycXFxF86KEBKvsAiBMbQHheW/Vsr6crhMtCgXBSjH4/vmhSTA0ixLpT5dmjsrWW3zly6Nc4EQ62HJUjDw44XlFnmO99prr3bboTpejx49xGazqXYXX3yxVFZWerWBxReMHDnSa3tRUZHyE9b2a21HjRrV5jjathUrVnT53Agh8Uk4/WKjGSiiUIAraxpVUQtYXLHEenKSLaSWxK766+rTpeH6gtYgUm/3C84FQqxH0HcWBM8hUO3ll1+WTZs2KSusHrx+8uf3a4QLL7xQWWovuOCCdtvAjQGfESNaK+3Amvzwww+rADj4CGtBeHCZQCBdenrbvItwxdBX2cO/sc1fO21/R0o8Slj7pqYD8KnetWuX3+/V1tZ6LUnoocwjQzzLvaHZpUr5onoZCjaoQgtZyUpZWrZhm/Kv7syVxF+fNY0uyUi2dfjdaJM7zivRWS8DcmxiS7BJc4tbkhKTxOUWSWiuk7KKnQHLKpzsW5wq7qYc+WVrs4hLZECuXfYuzpFxPVPU34FQzAUSnXM9Vqi1oNyhd4VNIUaA2oMPPihjxoyRww8/XBwOh5jBW2+9Je+++678+OOPnhQ4/rjiiiu81jGGffbZR0466SSV8UK/v6N+fPcF0lbPk08+qTJX+GPRokVSWlra7ne1NiS8UOaRIV7lPgCfLETv/vbgjDR08DNFsGGoiSa5D8KnyM8OV7Us+KpEooHe2tK1TRpKtsmXJdaZC7FONM31WGKRheReUlISPoUYlmEoxVqgmlka/UUXXSSXXHKJyiyBbA/6oh9YT0pK8mvt1arlYd/ChQs92/Lz85X1uq6uTqVR0wP3inHjxnm19WcF1tww/FmP9Vbtk08+uY2FeNq0aTJ+/HgZOnSo3+/hiQqTCG3aOy9iLvEgc6PWw3ASD3Lv6HrMWbhJvfqHL6lWilflZs5LU5k39Nepo+v3zdpy+WLNDqlrcklWSpLsamiWNIdNJu1dKBMH5AcldyvNl0BlZfaxzZJDZdUuWfrj9zJ6n3GSl5NlifOLdeL5HhNJai0o95UrV4ZPIYb/7mGHHSZmgnLP27dvV5ZnfHzJzc2V4447TgXCtQduLFr1PL3v8LJly2T//ff3bIfFFsfTXC60tmjni7ZN39YXpGjDxx9w30Dxko7AJOqsDTGXWJR5NATrxKLcOwNnO6Jvo5Ts3iZLtzfqUnulysh+RZ40dJ1dP2Ql+LG0UcoaEmVYUU5rAQm3W6UL+2Fbgxw0PK1d/1d/crfifDEqKzMxUw5aX8s3litL8NvLymVEX7unr0icX7wRj/cYK5BuIbn71q4IqUKMfL5IQ4aUa2aBALl58+a12Y7gOvgIf/jhhyoQrj2Qmg2WYH0qNuQQTklJkeeff95LIcY6/pjAgqu3MMPSi/PS2jqdTpXxAuuwWhNiZbRgHeRwxR9aWJ2QXxVzHVH0JHIYSe3V2fUzUlUtkIAwq86XQNKgmYGZctD6go8w3CJgCYbyq+8r3OdHCAmhQvzoo4+qfMN4IkCRjo6C0YwCxRU5gn2B8opMEto+BPGdeuqpcsopp8jAgQPVjQUK8yOPPCLDhw+Xs88+22sMSNuGwhz4t1aYY+bMmaqdlnINnHnmmaoQB1wfoITD4gvf4NWrV6vCHIRYmXAXNCDmpvYycv30ab2wT3vd7pvWK9rnSzgLNpgpB31fCJiDjzCuCyzB+r5YkIIQ6xH0HQZmcWSCQPCab4CbPhNFKMCxu3fvLg899JByscBx+vTpI5deeqnKF+zrw4J8w5mZmUrZfeCBB5QlGv7P2K4H2SiQpQJFOODHDGszggZhmWaVOmJ1zLYekvAWITBy/ZASzEhVtViZL+Eo2GCmHALtiwUpCLEO9q5UkXv99deVTy8CxszKMuEPWIjx0fsSv/322wH1AWUZn86Aov3CCy8ENU5CIomZ1kPiTXlNg/xa1aDkWpCR4lc8Xa06pr9+3TId0uRyi8OW0Ob6mfW6nfMlODl0dJ31fTWnJ6ltzU6XVNU3h/w3aGT+sTIeIe0T9K8TgW333HOPXH311cF2QQgxEa0ogBnWQ/KbAvH4Z2vl67XlUtvolPRkuxwwsEAuPnSgR55mBWShvzG9cmTJ5irlg5qYkCAtbrf0yE6VfXrneI5n1ut2zpfA5GDkOqPtaHUNd8qXa6qlfz+RL9eUS1JqmozRXUMzMTIuKwZPEmI1gv51Iv0Z8v4SQqwDg3XMBcrwf5ZslQZni6Q7bLK1ql6tw4p4zdTBpgdkuT3/SvBa/rZdTH3dzvliXA5Gr/MeRwndVcMywXNFzcbIuCIVPEmLNIkmgr6bnnDCCfLxxx/LlClTzB0RISRoGKxjrpsELMNQhvvlp6l0ji0tLbKhok6+WrNDzjigj1JIzQzIWrq5SjJT7DK2dw9V3tlhT5A1ZTWypKRKJu3dzXQLI+eLMTkYDbxDuyV7ruHwvbqJODfJQYO6yYryZvmxpEoONvkaGhkXCHfwJC3SJBoJ+lfwpz/9Sc455xxpbm5W2Sb8ZZQYO3ZsV8dHCAkCBut0HfgMKzcJh82T2xxLrGM79hdmpniCqJwtbmlsdklyUmKXA7IcdptoXwtHkBvniznBcvp2SfZEVXkuyW6TnNSEkFxDI+MC4Q6etGo6P0I6IuhfgWYZnjVrlkrBpgcBCZj4ocoyQQghoQZWNPgMw00ClmHNQlzb5JLiHIfarymSy3+tlPXlNdLkbBGHPVEpx+P75gUURMUgt8jRmUXT6LXRt3NnJattoQxsNTqucAbbWjmdHyEdEfSsfO6554L9KiGEWB5kk0AAHXyG4SahLMNNLkmxJ8qBg7p5sk1AAa6safT4GWttkpNsAf3hZ5Bb5OjMomn02ujbqVLMWaKWqEIXisBWo+MKZ7BtNKTzI8QfQc/K008/PdivEkJIVIBsEvhjDp9huEnAMgxl+KLJAzzWMFgX8zOSxZYgKlVafrpDXG6RJqdL7Q/kjz+D3MKPUYum0WujtVu2YZtymeidl6ZKMoeqCp2RcYVzXvFNB4lW+JhGCCHtAEUI2SQQQOcvDzGsXVCUBxVmSmFWsjQ6WyTZnihluxqlttEVsDWMQW7hx6hF0+i10dqN7u6QBV9tkukT+khhfm7Ixm9kXOGcV3zTQaIVKsSEkHZh2qTf/sgjgK6jQgxQlpOS7ab4Z4Y7yM2q1zkc49Jfw5w0u9Q0uiQj2dbuNbTqtTEyrnCNnW86SDRinTsfIcQyMG2SMTlEuzXMqtc5nOPCNRpZnC2frSqTRRsqVTEUFEXpnpUix+9THPA11Ma+fGOpwLFmzsJNMqJvY8Bjt+q1MQLfdJBoxNp3a0JIRGDaJONyCNQaZiVrrFWvc7jHtWxrtVTWNkmzs0VsiQlKGcU6th89qmdQY5fmehmQJVJSWSclu7cFPHarXptAYDo/Ek1QISaEeMG0SYHJwag1zGoWP6te53CPCwVYFq6vELgPD+uZKS3uBElMcMumynpZsK5C7df7jRsd++juGSKNO9QD0tLtjQGN3arXhpBYJui78Pfff2/uSAghlsBosv9YJ1A5QEHpkd3Wz9jX4gdLH6rQYYn1+at3SCSw6nUO97j0BVjsNhRESVRLfQGWcI/dqteGkFgmaIV4v/32k4kTJ8rLL7+sqtURQmIDLcgIvrAIEANaoBhK0gYbKFZaXe+1DDWwspVWN6hlV+WAFGqw1mEZjBx8LX7FualqiXVY/IIdoxWvc7SNSyvAgvzRKLwCtAIs2I794R67Va8NIbFM0Arx888/r24ap512mvTq1UtuueUW2bJli7mjI4SEHS1QDIFheEW7dWe9WgYbKFZd3yQXzPleLvvXErWOJdaxPRTANeG/K0pl1idr5NFPf1FLrGN7IOA8x/TKURY5WHI/W7ldLbG+T++cgORgRYuf2dc5WselFWBBMRUUYMEDG5a+BVgCHTveAIDWwhyBjd2q14aQWCZohfgvf/mLfPvtt+pzxBFHyAMPPCD9+/eXE088UebPn2/uKAkhYQV+rQgmgv8jXiFjifVgEvlf/9Yymb+6THY3tip9WGL9hreXh2Dk5romtNrmQILX8rftEtUWPzOvczSPCwVYjttnLynOSVX+3FhiXSvAEszYUZADYBnM2K16bQiJVbp8F4brxIsvvigPPfSQzJ49W/7+97/LlClTZOjQoXLJJZeoinYpKcafsAkhsZM2aXNlrXIJcLa4JVcpfU613F7vksUbKtT+Xnnppo3bzGAk9LV0c5VSWMf27qGUa4c9QdaU1ciSkiqZtHe3gC1+VkvPZtX0WOEeV2cFWALBrMIcVr02hMQqpoU2OxwOSUtLU0tYPurq6uSCCy6QQYMGycKFC806DCEkjHQWKNYZa8tqpNHpEoct0ctVAOvYjv1mondNgBIOxRjLjlwT2vM11vflsNskI8WulsG6OVjZ4tfV6xwr42qvAEswNDS7vJaxdm0IiTW6/Av76aef5IknnpBXXnlFmpqa5OSTT1b/huUY+84991w577zzZOnSpeaMmBASNQwszJBkO6p+NYvbvcfdwO2WJleL5DiS1H4zgRUNgVArfq2U9Tt2S5PLLQ5bgrjcIuP75Xm5JnSWBs23Ch0U+a5UoaPFz7qYmRIPD1aPf7ZWVmwslROKRG6Zu1yG9+2h3DKo1BISgwrxa6+9phThr7/+Wrp16yZXXnmlsgj36NHD02bUqFFy9913y9SpU80aLyEkioA7BFwC4DO8c49FFUt7YqLs1y/fVHcJoOUFrqhplAZnS2vqrCaXCpCCdVevkHRW+CBUbg4sVmA9zCyCAWX4P0u2SqbNKVIkUra7UdYt2ar6glsGISTGXCb+9Kc/SW1trTz77LNSUlIit99+u5cyrNG3b1+ZPn16V8dJCIlS7j1xpEwe0l0yk1sVSCyxfs8JI0w/FqxzTc4Wyc9IVsFMsPRhifXGZpfHLcJoGjQruzkQczAzJR6KeHy9tlw9jBXvSdeGJda/WrND7SeExJiF+IsvvpADDzyw03bIPPHcc88FexhCSJSTneqQJ/88Vn7ZvF1W/rBQZp0yRvbuFZrSs/DrhQIzqDBTCrOgBLdIclKilO1qlLoml9oPC62RNGiBVKEj0YvRuRBokY/ExFZ7E5b6Ih/BBusRQixqITaiDBNCiEaP7FSvZSgKaujTm9kTE1QgHJa+6c1CkQbN6LhhJfxpSxWthRbBzLmgL/LhdLUG02EZTJEPQkh4Ceiuf8cddxhuiydsFOsghJBwBTYZ9fs12s7ImIyOWwu2wit1ZUVMtquCEAy2iixm+orD+juhf768+f1mWVdWK9Jd1NLtTpKJA/JpHSYkVhTimTNneq1rUde+2zSoEBNCwh3YBEUU34H/J155w+8Xio2v36+RdkbGZHTcWrCVFuy3taperTPYKvIYnTNGGFmcreZES1OrhTjJnij5jmS1nRASIwqxVucdrFmzRn7/+9/LWWedJaeeeqoKqCstLZWXX35ZBdp9+OGHoRgvISQGMbOghlG/387aGRkTMDJufbBVv/w05VeK+ylKBCPYCgUh6FsaOczyFcecWba1Wnrnp8ngvEyRls0ydXh3WV3pkp+2VCvFmz7ohFiToKNDLrvsMlW++YYbbvBs69Onj9x4443S3Nwsl156KZViQkjYA5sCTW/WXjsjYwJGxs1gq+igqynx9HMm1ZEo0iCS6kiSnFRb0POYEGLxoLovv/xSDjjgAL/7sP2rr77qyrgIIRagKwFuoQxsCse4jIzJ6Lj1wVbamzYsYzHYKlxzxoro50Ozs9VlAsuuBGtalXi+ziQ2CfrXmZycLN99951MmTKlzT5sRwlnQkh0YmblLiOYGeQW7jEZaQN3CATQwWcYbhL6giEHDuoWE+4S4Z4zVgTXe0yvHFmyuUq+XFMl/fuJfLlmhySlZsg+vXNiwjrM60xilaB/nccff7wqxpGRkaF8iHNzc2Xnzp3KhxjZKP785z+bO1JCSFRW7jKKWUFu4R6T1mbBunKprG1ShUAmDihoE5CFbBJoB59hZJkoznEoZfiiyQMkFojEnLEiv4WZawHme0qWS2zA60xilaAV4oceekjWrVsnl1xyifIXttvt4nQ61evCgw8+WO0nhMR3gFsgmBHkZva4jAZb+Wbb8V0H+B5K9yKADj7FGHssWIYjOWesKIelm6uUe8TwvQpEnLVy0KACWVHeLEtKqmTS3t2iWg68ziSWCfqXmZmZKZ999pl89NFHMm/ePKmsrJT8/HyZPHmyHHHEEV7p1wgh8R3gFghGgtycLW5pdLok2Z4YlnF1FGzlazErqaxTluL2LKNQgsOpCDc0uzzLrHbaIAtGV5R0/bWpb3bJ7nqnZKbawzZnrIJeDki3Jk6kXbNJTmpCTMgh0vcGQkJJl2fukUceqT6EkNgKDMIrbyhIWr5x+MbCZSBSgUE4LgLQVvxaKet37JYml1sctgRxuUXG98uLyLisbDHTfD2XbywVOGXMWbhJRvRt9PLpNatYCGSfYrfJ1+vKpL7JpY6NY6Q6bDJlcGFMBZMZ/e24s5LVNiv8dmL93kCIGXR59n766afqU1FRIQUFBXLYYYcpKzEhJDoxs3KX2eOCklVR0+gpbqEFpjnstoiMy8oWM81yLc31MiBLlOW6ZPc2L8u1WcVCcI7bdjVI2a4GZb132BLVQ8LuhgQp3d0YN1ZD/W8HSmPvLFHLjOTUiP52Yv3eQIgZBB3629TUJMcee6xyj7j33nvlueeek3vuuUcpxH/4wx9ULmIz+Oc//6luzgje8+WHH35Qx8O+nJwcOeGEE2T9+vV++3nsscdkyJAhKjtGv379VECgvzGWlZXJjBkzlHKflpYmEydOVAo/IfEErIhHj+qprD4Oe6JaYj2Yyl1mAWtmk7NF8jOSVeAaLFVYYr2x2RWR9E+BpouLhOUa1w5giXVYrrHft1hIj+xUtcQ6Av+w3yhou3VnvXpgyUtLkjSHTS2xvqWyLqC+YuW3g7kJsIz0byfW7w2EmEHQd2tkkvj444+VMgwFslu3brJjxw554YUX5KabblL7//rXv3ZpcFu3bpWrr75aevbsKdXV1V77Vq1aJYcccoiMGTNGXn/9dWloaJBbb71VDjroIFmyZIkaj8Zdd92lykhff/31SoFfvHix3Hzzzar/2bNne9o1NjaqNHJVVVUya9YsKSwslCeeeEK5hHzyyScyadKkLp0PIfFWuctMMA4ocoMKM6UwC0pwiyQnJUrZrkapa3JFxBprVYuZEct12e6GVjcJh01VzgNYKst7o1P5FBv1J0ZbXJvctCTplpkirha32BITZMfu1u2B9BUrv53R3R2y4KtNMn1CHynMz5VYwYr3BkLMIOhZ/Oqrr6qqdNdcc41nG5RQKLA1NTXy4osvdlkhPv/881XGiry8PHnzzTe99kH5hbX3vffek6ys1lCRcePGyaBBg+SBBx6Q++67T22DK8edd94p55xzjtx9991qGxRpWIehFF9++eUybNgwtf2ZZ56R5cuXyzfffKMswwDuH6NHj5Zrr71Wvv322y6dDyHxVrkrlP6LSSn2sPkvQqlr74+/kdRs4R6TEV9Wu621WAjcJJwul7S4EyQxwa3cUJASLpBiIVrhEfRVKG5lOdQKjwTaV7hlRaL/3kCIGQQ9m7ds2aKssf7AdrhPdIU5c+bI559/Lj///LNSXPUgvRsUYZSO1pRhrXQ0FNh33nnHoxAjCwasx2eccYZXH1iHJXvu3LkehRjfGzx4sEcZBkgnN336dKX8w6JcXFzcpfMihESPNdZIEYJwW8yMjMmILys+E/vnyxvfb5Gff92tLLqw7KYl2+V3A/IDsuhatfBIJIpIGAlmJIRYj6Dv2rAGL1u2zG+lOmzXuywECvx4YbmFO8Zee+3VZj/yH9fX18uoUaPa7MO2//3vf0oJTklJURZfMHLkSK92RUVFyk9Y2w/wb39KvnacFStWUCEmJIKE2xobSBGCcFnMjI5Jk9UPa7eqqhBF2SkydqC3r+fw4mz5bFWZUuJa3G6VKiw/3aG2B4oVC49EooiEkWBGQoj1CPrujcA5uC307t1bBbNp/Pvf/5aZM2d2qVLdhRdeqCy1F1xwgd/9cIMAcKXwBdvwahBV86D0oi1cK9LT0/221frS+m2vT/1x21Pi4UOtZ+3atWoJF5Jdu3b5/V5tba3XkoQeyjy65b7/XqkyunsPqWl0SUayTVKSbFJfWyP1Yi7I2wsrHxSb0d0zWlNMZSUrpWrZhm3KRxTHDieBjAnZHlwNteJwtwYPY4n13btTxJ6YoPpas6VMhnVLksMHZUpdY4ukJSfK1qoG+WVzmZQVpwZ8fudNLJL/G50vZbsbpTAzWXLSHOJsqJNdEYipi8T10x9zSL5DpFnUclVFfcTmTDzBezvlrgG9K2wKMQLVvv76azn55JOVstmjRw/Zvn27GgSssdgfDG+99Za8++678uOPP3Za3KOj/fp9RtsF2lbPk08+qTJX+GPRokVSWlra7ne1NiS8UOaRIZrkDtsmrHzS+NvDLtwPUHABAVPRMKb+nmWpNG4ulS83++mrBb4frcu9TTq/rRKf189zzD1JjIqat0pRhOdMvBFN95hYYpGF5F5SUhI+hTg3N1ed/PPPP68q1cF6OnbsWOVCAd9eWGUDBcr0RRddpMpBI7MEsj1oKd4A1pOSklRFvPYstqiYp6Kpc3LUOtrCfaKurk6lUfNti0A8DbRtr0/gz3qst2rj4cDXQjxt2jQZP368DB06tN0nWsgRbfxZsYn5VFbtkqU/fi+j9xkneTnt1e6yJrBA6S2j0TQuM+UeDjngGPD/xCtvuGZoRQiUT25emsoeEMpj+zs/o2PSt4OFEkrZtqRiWVXR5GkHtDZ98tKk2eWWJFuCbKqsC/n5WUlWRvoKZCzaMQfm2qWXa5tsthXJ2p3OqJJptMK/p5S7xsqVKyVQuuTwBqX3vPPOUx8zKC8vV1bmBx98UH38KeHHHXecyjiRmpqqfJV9wbaBAwcq/2G97zC277///p52sNjieCNGjPBsQ9v2+gT6tr4gRRs+/kCeZH3wnz+gDHfWhpgV7FKurDhvLyuXEX3tURHsEongILPGZabcwykH/BoRDAX/z6XbG3VBfKkysl9RSFJpdXZ+RsdUV90g5Y2JYk9OFyfk0izitCWLPTlJbW+xp0qP7BQZ0rtBvt2yQX7ctlMSE0Ra3CI9spNl6j7dLZ8qzCxZGenLKDimJtOSil1yej+Rz9bukqTUtKiQaazAv6eUe4af2hWdYVoECHxkf/nlFxVMh2wPwQC3C1ibfUFwHTJOfPjhhyoQDpkfUBTk7bfflr/97W+SmZnpMZHj+1dccYXnu8ghDOUYlmy9Qox1WAxgwdU4/vjjlaUX6dW0tshogYwXWIfVmkQv0RzsEongILPGZabcwy0HKwbxGRmT0RLCezIUY6/XsmNnNWtglqyM9mUUb5lKVMmUkHgmYIX4/fffVzmI4bqAwDlUioP/LPL0IvMDQJDdK6+8otoEAhRX5Aj2BcqrzWbz2gd/3f3220+OOeYYVXBDK8wBhfmqq67ytIObA9K2oTAH/q0V5kDg39lnn+1JuQbOPPNMVYgDrg9QwmHxxbmtXr1aFeYg0Yu+chcCbOBTiD+OsBzhjyVSZlk1p6Z+7MOKstQfaeR1ReqxSI7dyLiAWXKPhBzCmVLN6PkZGZORtGs43pLNVaqi3tjeRdLkQsnlBFlTViM/llTJwXt3i/rfhBFZmTmv9DIdvlc3EecmOWhQN1lR3mx5mRIS79gDVYZhmYWrhMPhUJbTxx9/XPn8QtFEejIEw8Fy+9RTT8mll14asoGjDPP8+fPluuuuk5NOOklZjQ899FBVlMM35RvyDcOKDGUX+2GJhhKN7XpwXijTDOUe5wS/Y1TCg2WaVeoiT1eS6xup3BXpP1Qob4uKXvhjrM/batWxGxkX0NogFRfAsqOxt3edrSqHzsZtFLPPT7OOIrsBArrgwwpXAc06qj+ew24TrWuryNNMWXWUEs9Muev7Qgo7yD3JbpOc1ATLy5SQeCegXyaUSQTNIQsErLmXXXaZck8499xzlQKscdZZZ6kSzmYpxLAQ4+MLAuKMWm4xFiPj6d69uxo7sQ5m+PcZfYUcKUXq8c/Wytdry1tL6SbbVZED5HXFH0/fCm1acFCkx250XI7ERPnfmm2SLs0ydITIR8tLpVaS5OjhRV5j79QnNAJyCMRHusv+pwbPz+jxOishbNV5ZQQzxx6qvqx2nyGEdExAv04UpvjHP/7hCVhD2ebHHntM+d7qOfHEE1XVN0LMwAz/PiOvkCMFlGFU+GpwtqgKXyh/i3Wc3zVTB0ekQpsRjI7rp1+rpaKmSewprT6Vjc4WqWhokmXbdnmNvbPrHAk5BOIj3VX/U6PnF+jxtKwGvtkNrDqvjGDm2EPVl9XuM4SQjgno14msDHA30ND+Db9dPUhfVl1dHUjXhPjFTP++zl4hR8pNApZhKMP98tMkMTFRWlpaVPlbVPw644A+yn0i3MFdRulsXJsra2VzZZ0KKEq2t76OxhL/KqnAvlrplZdu+DqHUw6B+kib4dfc2fmZ7UcdqDy76hZiJmbOhVD0ZaX7DCGkcwK+oxkteEGIGZjp39fZK+RIAJ9h5SbhsCllGGCJdWzHfijE4QzuCoTOxrW2rEYanS5lncxMxfk5JTM1SVIaW9R27IdCbPQ6h1MOgfpIm+H329n5me1nbFSeVkz7Z+ZcCEVfVrrPEEI6J+BfPDIuIIANuFwutVy1apVXG991QoIlFH6O7b1CjgQ4J/gMw00ClmHNQlzb5JLiHIfar6ej4KBI0t64BhZmSLLdJlX1zSLuPQ/Qbrc0uVokx5Gk9gdzncMhB/2YumU6pMnpFoc9oc2YQuGH2975hcrvtzN5WjXtn9lzwcy+rHSfIYR0TsC//BkzZrTZdtppp3mt4wZN6zExg2j2czQCrL8IoIPPMNwklGW4ySUp9kQ5cFA3r2wT0Qisv7hO81eXyc49FlUs7YmJsl+/fLXfqtcZxxzdK0eWbN4pHywr9SpcMaZ3jmdM4Rx3JORk1bR/hBBiJgHdxZ577jlTD06IEazqP2sWyCaB84PPMNwkYBmGMnzRZNR1i37uPXGk3PD2ctnw645Wl4lku4zq103uOWGE5a+zkcIV4R53uI9n9XR3hBBiBgHdxU4//XRTDkpIIFjVf9YscC7IJoEAOn95iKOd7FSHPPnnsfLL5u2y8oeFMuuUMbJ3r+6Wv85GC1eEYtwdBa+FW07RnJ6NEEKMwjsZiRqs6j9rFlCCY0kR9qVHdqqs3LOMhuscaOEKM8YdSPBauORkRXcWQggxm8iEBxMSQRqaXV7LWAQWxtLqBrXsKOXbT1uq1DIcxzMqdyN9GQEp3eat2q6WXbGMQvmDRRRollFYjUNhGdWC12CNRRAfllifvxruJpGb71DIjx7VU6UQc7a0qCXWY8VtiRBC+GhP4gbN+rZ8Y6nAO3fOwk0yom9jRFNHmY0RC2NnlfHMPp5RuZuV2qu6vkmuf2uZ8rFFajdkuYAlE77McN+wqmXU7OA1s+e79lDQ3johhEQzVIhJ3KBZ36S5XgZkiZRU1knJ7m2WSB1lFkbSY3VWGc/s4xmVu1mpvaAMI6sFFEIofnVNTWodgX3wZQ7EXzecAWx6Fw1ni1sam12SnJTYrotGZ0UyzJzvvtcGfVXWNsXUb4cQEt9QISZxgd76Nrp7hkjjDqXcLN3eGDOpo4xYGNHGSGU8s44HjMjdLOso3CMWb6hQeY6Rug5GzGR7ojrfRevLPZXxjFqkwxnApvW9/NdKWV8Ol4kWcdgTlXI8vm+ex0XD6FsAs+Y7064RQuKB6NYACDFIPKSOMnKOZbuNVcYz63jAiNzNuj6ofIc8zsiO5mppzYfeguTBblHbtcp4gVqkwxHAhv6hAFfWNHqs91pO6uQkm+f4RsZt5nyPh98OIYTEhuMkISEIkDIruMtK56hVxoOiBcsw0CrjYbtvZbyuHs+o3M0KYCvOac1gAR0YhTQSExI8BTX0+/VWz0GFGZKdlqSWWIcFNZTXvL15hXVYf/MzkqV3bqqSB5ZYb3K61H6j4zYzIDASwYWEEBJueCcjcYE+QAoWtd5ZopYZyaltAqTMCu4KN0aCwPAxqzKe0aAzI3I3K4AtK9Uh+ekO2VrVIHXNbkkQl6ekRn6GQ+0HsGpW1zerc1+0AcF3Lcq1AhZabA+F1bOzeYVjwko/qDBTCrOSPWMq29UotY0uj8XdyLgDme+dwbRrhJB4gAoxiRu0AKllG7ahYJpKHTWyX1GbACmzgrsigZEgMDMr4xk5ntbmh7VbletCUXaKjB3YNmVXIAFs7QWUYX3vHpmyo6ZRpS2TPcqww54gg7tnelmkERS2YUeNpCQlSmqSXSpqG6WhuUWyUpNCmlINiivGvK26wWte+RbASEq2+y2AYXTcRue7EaxYRZAQQsyECjGJG7QAqdHdHbLgq00yfUIfKczPjakAIiNBYGZWxjMadGYkZZeRvgwFwiUmSm6aQ5JsCQIbMf7f7HKL3ce6r5Vf1goxexdkNhfMq8UbK2VDea3YEkR27G5UVe+g3C7aUOGZV51ZydGP0XEbme9GsVoVQUIIMRve0QjpQuqrcIMiGkaUWCNBYNhfmJliyvl0dDzNMtrcUCdDclvTf22raz/9V0VNowp+G1iYIWl59oCs97g+uWlJ0r8gQ+Ug1lwKkIs4N9XhFcTn2y4v3dGmnVmgv9Wlu5UCbLclSIrdJjubnOJ0udV27XiaJXbBunLZWdukrLoTBxR4LLHBjDslyea17ApWqSJICCFmwzsbiRuMFCowmvoq3IS7mIaZ4164vkKWbK6SNGkSyRXZsKNW6qRZKXJ6i3tnBTWMWO9xfbLTHJJW2ySje2Urtwm4S6wpq1EBaHqXCSPtzMJua/X9rW92Sb+s1nR3GS025ceN7div4QlcU97P3tb0cI+bEELiBd49SdxgpFCB0dRX4SbcxTTMAhZLKMTovzClNatFbZNTyhpa5NsNlV4WTa2gBh4+HLZEqapv9iqoYST9V4/sFI/bAZTE9oLz9O4JHbUzC6dLJDs1SVmIK2qbldUa1t3UJJvajv1GCmCEe9yEEBIvWDdknhAT0VsXERAEsPRNV2Uk9VUk3CT0xTR6ZKeqJdYRGIf9RvG1shbnpqplqNKNNbtcyl8WSm5Wyh7rbIpdrZftalD7AQpm4PjYXpCeJDlpSWqJdRTawH6j6b9g6T56VE91ffFwgyXW/QXxGWlnlI7S9Klgv+6ZyiqOc7MlJqgl1rEd+9ukVEv1n1LN7HETQgihhZjECUaLCxhJfRVuKxx8hsNZTMPM89tZ16yUPwS4QYEHWGLdnpig9vfKay2oATcJWIb154h1bNcKahhJzWY0AMysQDEjLijod3y/PCmvaWzNMpFkl7pmp1J69+/f6jYCZbq6rknqGr1TqsF1pLrut5RqDHAjhBDz4fs1EhfoU1q5s5LVNn8prYymvgonWjENuEk4XS5pcaPYhFu5ciBlWjDFNHB+3TIdHh/Urp5fe2nQMLbCzGQ1VsceP1koudATu2Ume8aOADoofnCTQKEQraQ0SjDnOJLU/lCl/zISKNbe+QXiguI79r1SUr3Gjr7xgADfdbhSpDhsyl0CfseZe4qcBDpuQgghxuDdlMQFRgsVWLEIAay/E/rny5vfb5aff92tLK4oS5yWbJOJA/IDLqYxuleOLNm8Uz5YVuqp4tYjO1nG9M4xPUAPY/vdgALZWrVFahqa1HdqGpzisCGgrsAzdlh/IV/4DJfXNiulGcowrMj79cv3lFs2KzWbWecXSJo+I2PXwudUQJ36r216OkIIIeZDhZjEDUYLFXSW+ioSjCzOVopZs6tBWtxuSbInSn56stoeKFr22lb167dlMFl4jVhHhxdny2eryqRljw+sGrvDobbrQTYJBNDBZxhuErAMQxm+54QRQaV5MyNo0Eiat0BdUNobO9rCp7h/twyV7g8uE7jGcJtAXuVIp/wjhJBYhndXEjcEUqigo9RX4QZWyGVbq6V3fpocvHeB1DS6JCPZJhsr6uSnLdVKgTeqKKEvpEBDENrY3kXS5EJGh9a0XT+WVMnBe3cLqK/OrKNg+Z6xD87LFGnZLFOHd5fVlS5ZtqVaDtWNHanVkE0CAXRaHmLNMhyIrMwqrGI0zZvexQZtgnWxUSnVUpMkzWGT0Xtle10bbGdKNUIICR1UiEnc0Vmhgs5SX4UbvRUyPTlJfUAwgXD6vhx2m2hf62pf7VlHgdYm1ZEo0iCS6kiSnFRbu8eDEhyoIhzImMw8P32at6662DClGiGERA6mXSNxR0Ozy2upx2jqq3CiTzeG1G8YC5a+6cYC7auj1GVm9aVvU9/UrNpgGczxgpJVgzmy6kjuZqZBY0o1QgiJDLQQk7jBSKU6WP2MpL4KJzjeGBUIV6Us14kJCcqPGPmI9wkwEM7MoEGjfY1QPsTbZeO2etl7qMjHK7ZLgiNVpu1TbLoszQwaNCp3M9OgMaUaIYREBirEJG4wUqku0NRX4eI3D+YEr2Uwns1mpi4z0teKrdVSWdssGQmteYibnS1S09ysth8zqqeYjZlBg4HI3cw0aEypRggh4YUKMYkL9K4Qo7tniDTuUMrb0u2NbYKtrJb6CmNf6gmE6+HJHYxgqyUlVTIpgEC4cFs0UUXvm3XlKoVaRga2OyUjxS6VNS2q+h72B5I2LtxBg2bKnRBCiHXh3ZzEBfoAKeTwBVj6q1QXitRXHRV26KxNMIFwRo5nJu1ZNFFFr2x3o7hUkY3WbVCOXS2iSjoHUmXPCGbKKtC+oNzjfJBtwsxzshLhnleEEBIueEcjcYH2B3z5r5VSWeWSft1FftpSJeVNNhnfN8+rUp2Zqa+MFInorE0gqb3MOJ6Z5KYlqQePZpdbUux7snvYE6XZ5RJni1vtNxMzZWW0LyiJj3+2Vlm8VYntZLsqOnLxoQNjRmkM55whhJBIEBt3a0I6AYoJMgBU1jRKs80p0l1kW3WD7HbZJTnJ5rdSHZTgrlaqM1IkorM2gQTCmXE8M0my2VSJZhxrV0Nrhg4s7YmJUpiVovabiZmyMtoXlOH/LNkqDc4WSXfYVIltrKOfa6YOllggnHOGEEIigaUe7ZcsWSJHH3209O7dW1JTUyUvL08mTpwoc+bM8Wo3Y8YMdSP2/QwZMsRvv4899pjal5ycLP369ZPbb79dmptbU0DpKSsrU30XFBRIWlqaOvann34asvMl4QNWPFi58jOSpSgrWW3DEutIpaVPp2ZW6ivfwg7FualqqU/hZqSN0TGZeTx9n6XVDUGnm4MVFWWnu2elSPoeBRJLrO/f7zfLvJkEKquO0ut11hfcJGAZhjLcLz9NZaDAEutfrdmh9kc7gc4ZQgiJRixlIa6qqpJevXrJn/70JykuLpba2lp5+eWX5bTTTpONGzfKzTff7GkLhfmzzz7z+j62+XLXXXfJLbfcItdff70cccQRsnjxYtXP1q1bZfbs2Z52jY2NMmXKFDWGWbNmSWFhoTzxxBNy5JFHyieffCKTJk0K8dmTUAK/R7zOHlSYKXtl4B1wtYzqlStbakRqG11e/qBmBZ0FWriio0ISRsZk5vHMekWOvqAQI1NHc0Mtsj9Lv27pkpSSrsphh8KlwKisjKTX66wv+AwrNwmHTRITW+WCJdax3Wwf6UhgZrETQgixKpa6ix1yyCHqo+eYY46RDRs2KOVVrxDjj86ECRM67K+iokLuvPNOOeecc+Tuu+/2HAPWYfR1+eWXy7Bhw9T2Z555RpYvXy7ffPONsgyDyZMny+jRo+Xaa6+Vb7/9NgRnTMKF3h+0zx4LsS0xQarqG9stsdvV1FdGfVADKf3b0Zj0x+uW6fD4PwdzvEBfkXcUbKWlZlu2YZuIs0IdZ2S/oqDSvBk5nhECTa/XntwhQ/gMw02i2ekSt0rv5pbaJpcU5zjU/mDGbqXgNTPLUxNCiFWJijsZXBjgzhAoH330kTQ0NMgZZ5zhtR3rN910k8ydO9ejEL/zzjsyePBgjzIM7Ha7TJ8+XW688UZlUYbVmkQnen9Q/GHvnSVqmZGcGpRvcKDH7MgH1cxCGUYKSXR2PN9X5FCAoAihvW+KOiOWZM3KOrq7QxZ8tUmmT+gjhfm5QcnUzKBBM9LrwfoLC/ib32+WlaW7xZaQIC63WwVlwgKutw5bLeAxEsVcCCHEqljyTtbS0qI+O3fulDfeeEM+/vhjefzxx73a1NfXS48ePWTHjh1SVFQk06ZNkzvuuEP5HWvA4gtGjhzp9V20h5Kt7dfaHnTQQW3GMmrUKLVcsWIFFeIox9tSKdI7L63LlkozCldobRasK1dWSowLylQw4zJSSKKz4wXyijwQS3JKks1rGQyBBA1W1zercSJ40reNmen1UIlv3uoyaa5uUA8gSYmJyjd9RHFW0GO3WvCamcVcCCHEilhSIb7wwgvl73//u/q3w+GQRx99VM477zzPfrgx4DNixAi1/vnnn8vDDz+sAuDgI5yRkeFxmUAgXXp6eptjQHHGfg38W69M69tp+zsCFmwo53rWrl2rljU1NbJr1y6/34OftH5JQsv+e6XKoIwCWfrjJjlhZIHk5aRKfW2N1If4mKO795CaRpdkJNuUQqg/JtKPuRpqJdndKGkJzZLsbl3fvTtF7Kg9bJCGZpesKtkuRWlumbBXrkp1lmRLkE2VdbJy03YZV5Sijt3Z8RKdLilIbpGS2jqxuxI8r8idjbXSMz1NEp31smtXkzoeymCj8h+Knah2WclKicNDByzCeuW3q3PdyPHATxu2yc6qarElIKMFFNQE2dkosnS9eMaEcyxMaZGGlBbp0zNdmlta20FW2K6do5ExrdlSJsMKkuTwgZlS19QiaY5E2VrVIL9sLpOy4lR1PKNjD0SeRjHrHtPZPCahkTsxDmUeGWotONehdwVKght/6SxGSUmJUjDxeffdd5X/8H333SdXX311u99566235KSTTpKHHnpIrrjiCrXt3HPPlZdeeklZk32BewQyTsCtQlO8zzrrLHnqqae82i1YsEB+97vfyauvviqnnHJKu8efOXOmyl7hDyj0yJxBCCGEEEJCr0deeuml6u3/8OHDo9dCDOVRUyCPOuootbzhhhvk9NNPl27d/L+iO/7445UleOHChZ5t+fn5yoe4rq5OpVHTU1lZKePGjfNq688KjHbAn/XY16p98sknt7EQw5Vj/PjxMnToUL/fwxPVokWLVBt/luxoB5YxvUXJClhJ5pDPnIWbpKSyTr2G1qyxys85L0352xqVm5G+gJHjwYq8eEOlrPi1WmXhSE+2yfCe2bJfvzyP1TrQsW8qrZD1K3+S/kNHSZ8e+SGRFdrcMne5qo5XlJ0sbjcs3G7ZVt0ohZnJ8tdpIyQnrdUaa+QcNarqmlSf6EP7fiAy0Lfrk5fmZb0P9NpE83yPJyh3yjxeqLXgPWblypUBf8eSCrEvEPLTTz8t69evb1chBvjjoaU+0vsOL1u2TPbff3/P9tLSUikvL/e4XGht0c4XbZu+rT+Qpg0ff8CFIyvL25/QF0yiztpEE1YMDrKizOuqG6S8MVHsyenitP0WgGVPdqvtLfZUycoylrYLZzKib6OU7N4mS7c36oKfUpWvNALZSgM43tR9suWg4e1nOzByPFBd3yTXv7VMNm4rl/P3Frn2379I36ICuffEkZKd+ptyadb5ueypsqOxWX79tbHVhdrdGthXkJsqjlT8Fn87787OsbMqdEZlgHZDejfIt1s2yI/bdgr0bVQQ75GdLFP36e5pZ6SvaJ7v8QjlTpnHC+kWusdorrOBYA3NpBPmzZunFN3+/fu32+bNN99UlmB9KjbkEE5JSZHnn3/eqy3WYX2B9VZvYV61apVXejWn06mKgkCZ7tmzp+nnFctowUGwbjU5W61cWJ+/2tvPOtoxo3AFHhag+GjeS1pKq8yUtum/OqOzQhL646EgCYK3sGzveFD6emSntBtgZqQIBpTh+avLZHfjnjzIjU61fsPbvwW1mnl+KUmJ4nS1qEDCPfqwOFtaJMXRWo7Zl47OUatCp9KquVo8VeiemLcuIBmAPeGJXvktsK63RZtVFIYQQkgUW4jh84unC1iEu3fvrqy4yDLx2muvyTXXXKOsw5s2bZJTTz1V+fMOHDhQKbYIqnvkkUeUn8jZZ5/t6Q9uDsg3jMIc+LdWmAP+vminpVwDZ555pirEAbeHe++9V1l7n3zySVm9erUqzEGME0jarmjFzMIVZqa06qyQhNHUbGYdb3NlrbrmcE/IVcqoUy2317tk8YYKtb9XXrqpRTeQ+gyuBcgYgfa4VsgigXMNBN8qdHgoR/abDRV1qgrdGQf0UWnVjIwJvwnIHA8dY3sXeXJEozz4jyVVcvDe3QwXYCGEEGI+lrrTIgfwc889Jy+88IKqGAeTN7JJIDAO+YABFGYoywie2759u7hcLunTp49ynka+YF//FeQbzszMVMruAw88oFK1oWodtutBNgpkqUARjksuuURZm8eMGSMffvhh3FWp62pRgHiobGVmeqxAUlpBSUP1MzxgdFQBraMCHkZSswVKe8dbW1YjjU6XOPCQoCmkCQlqHdux31chNjL/2jsevpebliT9CzJUPyi0kZHsUG1zUwNLpxZoFbqOZK7/TTjsNtGatfeb6GpRGD3wX9aWXX2ZaaWCIYQQYiaWuqOhYIZvEQ1fcnNz5e233w6oXyjL+HQGFG0o4/GKWVbPWK9sZbYF3KiFsSNf1kDGvtRjqeyh3Fkc9lZL5ZKSKpm0x1JpFgMLM5Tyu7OpWWoaWhWzmoZmaXCK5KYmqf1mzj/ILiMlSWqadnv5g9U0OSUjQBcUfRU6WIY1C3FHVeis9JvQ5IlUbgP2BOzBRzkYP/5oiAkghJCuEN2aCbGk1TPWK1uFygLekVVQ82XF63tYKDVfVhz3mqmDgxq7EUtlV4H1t3d+ulRurpKG5lYbNJbI/tCnIMPLOmzG/MPY4XtbWdPokVVFbZOk2BMlOckW0LnB+ouHDsgZbhLKMtzkUn0dOKhbhxZ6K/wmNHkir/GALFHZKxCwF8xbDKsWDCGEELPgoz3xa/Uszk1VS6zD6hlo0FgsBweZHQgXqC8r/H2xxDp8WbHfqmPHvBm1V7YUZCQrn1mAJdZHFmd55pVZ8w/tYM1EpbjeuanqXLHEOoIHA53HsMAft89eUpyTqiyhWGL9osmwuYplfxN6eeI4AMtgfs9m3xsIIcSKRLepjljW6hnLwUHhtvYF6stqpbHj2jc5W+TgQd2kWzJcJjbLUSOLZEejTblraPPKrPmHdpDJoMJMKcxKVsF0CK4r29Wocg0HOo/RFhZ4BNAZ8d22ym/CzN9zPMQEEEIILcQkpJbDztJ2RSvhtPZpvqx4XQ8fVqD5smJ7IL6s+rEXZacoRQfLro69vfRz+nmVltw6B7D0nVeBpoNrD30/KK4BRR9LM+ZxYaY58zgcvwkzf8/hfqtACCGRgHcyEhd+v2YTTmufmb6sAC4FP5bsVEF0uxpgYa5XCg/OxexgK/28UhXXskQtUWxCP6/MSgdn9jyO1mAyo3IPtC/eGwghsQq1HBJU+i9ifnqsjlJawZcV1wY+w3AJQJYDKMPB+LL6Buht29UQVICePtiqur5Z0pLssq26oU2wlTavfli7VeV2g0V67MC2Fmmz0sGZOY+9zs/h//wiRWcp0DQ5LNuwDemfVflnVLwLRg68NxBCYh0qxCQu/H6tjBErpFm+rEaLTRhVyBZtqJT15bXKLWHH7kblPoKsDt+ur/BKP6e9atfwXTczHZxZ8xhjWryxUjaU1wriAdX52RKksrZJFm3wPr9wYtRqrclhdHeHLPhqk0yf0Cfo8s+8NxBCYh1qOySkVk9ibkorKKvBBnWZHaAHZfOX7btlZ22TUpiSkxKlqq5ZKWzYrgVb/Zb+q0GGZImysm5b5p3+KxTp4Lo6j3HM1aW7lQJstyVIit0mO5uc4nS51fZIBZMFmgINVfv0y67AewMhJFaxrhMcIRGmvUAxs48RipRW7Y1dH6DndLlUBggs2wvQ60gGdpsoVwJUg8tPT5KslCS1xDq2Y7/R9F9WDNzSnx+KiEChxFJ/fuGeV0yBRgghoYFmQEIiGEhldkqrzsYO6+/E/vnyxvdb5Odfd4stMUFcLW6V+eF3A/I91mEjMnC6RLJTk5QFtVwVv7BJg9MlqUk2tR37jZ6fFQO3MH4o43BiXldeJwniFrckSFJigtqO/eGeV0yBRgghoYEKMSERrMpldklfI2MfXpwtn60qUwoaMjkk2RMlP92htgfSD8Y2uEemspbCx7bJ5VYWVJdb1HZt7Nr5ubOS1Xp756cFbi1YV66UbASBTRxQELGgTowNyr2zpUWU0TqhdezOFrfaHoprE+9l0QkhJFLw7klIB6+koXBA8YDVEq/4zQ6kMtMyamTsYPnWaumdnyYH710gNQ0uyUixycaKOlm2pVoOHVKo2hiRAT779c1TwWZQilMdNqlvcinr8Ph+v8kpkPRfnQXfhRucO1wlUNzDYbNJk8ulin0kJmqZMMI7r6xoSSeEkFiAd08SNSBDQlerhVnxlbRZKa2MjB3og9fSkxNUZoj22nQmA9+x98q1txm70fRfvinckN0BluJIpTjDeealO6R/twyl0DY0uSQ3xSFp8CVOc0Ss2htToBFCiPlQISaWB8oIcuciXZjKkJBsV4UqkJvXbOU0Eq+kzUppZXTs6Hv5r0iXVqOC6qAQww1gfN+8Nm4OncnAyNiNpP8KJIVbuGiVp13NOTgSw1YNOSAAMdhqb2bMK6ZAI4QQ86FCTCyPbyGJrVX1QReSsPIr6a6mtDI6diialTWNHnlqVe+Sk2xt3ByMysDI2DtK/2U0hVs4wfEwlgqdrKCgQ1awrke62htToBFCiHlQISaWxsxCEkaJ5lfSnY0dllgomfkZyZ5AOATUIRCuyelS+6FohVsG+hRnPbKSW6+zo/U6dyXFWVeALGBBV7JKTPD8G1k5Gpt/k1U8zCtCCIl1qBATS2NmIYl4eCXd2dixDXIbVJgphVnJKkAMAWNluxqlttHlscSGWwZGUriFG5w3lF6PrJpblOUasqpr+k1W8TCvCCEk1mFhDmJp9IUkYBkGWLZXSMJMoKz0yE7pUGkJR/EOM8euL4ABizCUYyzbK4BhRAZmyEFL4YYgNqRug0UWS6zrU7gFghlj0mQFv+aMFLtadrVYiFGZEkIICR+8IxNLA+svAujgM4zX53qf1wMHdQtZtgkrFe8wEyhhI/bkIUYQG/IQJyYkSPesFJm2T3HASppZcjCawi3cY2KKM0IIiQ+oEBPLg2wS8L2EzzAsmsU5DqUMXzR5QEiPC8tie6+2w1m8w2yQh7iytlEpjraEBGluaVHry7fukmNG9QyoLzPloC/MgeC6HgUpQRXmCMWY6PdLCCGxDRViYnmgjCKbBALoQp2H2IiFMdzFO8wOUly4vkKNeWiPTFWKGCWJS3bWK0UU+43KNhRy0ApxtCY5C7wwh9ljot8vIYTEB9b8q02IH6CohcNFojMLYySKd4QiSDFJl7ohmCBFs+XgK/eSyrqAC3OE6towxRkhhMQ21nV2JAFh1eCuaMPXwlicm6qWWIeFEft9A9NqGjoOTLPSddYHKTpdLpVKDMtgghT1cvBYdvcUmwhUDkbkHu4xEUIIiR/41yHKidbgLqtixMKIDAGje+XIks075YNlpZKYINLiFumRnSxjeueExDps1nWG9XdC/3x58/vN8vOvu1U2B+TVTUu2ycQB+QFZ4M0MOjPLsstAOEIIIcFAhTjKiebgLititMTuHpVNebvql63brX2dRxZnq/6aXQ0qy0SSPVHy05PV9kAxK+jMzNLGDIQjhBASKFSIo5hoDu6yKkYsjJD7ks1V6hX82N5Fqtqbw5Yga8pq5MeSKjl4726myt3M64y+lm2tlt75aXLw3gVS0+iSjGSbbKyok5+2VCtlMhJBZ2ZadhkIRwghJFCoLUUx0RzcZWU6szDq5e6w20QTcajkHsx1bi9lnL6v9OQk9TFj7GYEnZlt2WUgHCGEEKNQW4pizHzNTIxbGMMt90CO15mvsZXnDC27hBBCIgU1piiGAUShl68/q2e45R7I8TrzNY6GOUPLLiGEkHAT+b9+pEswgCg65N5R1TuzjmfU15hzhhBCCPGGCnGUw9fM1pa7WenSjBzPqK8x5wwhhBDiDRXiGIGvma0pd7PT4nV0PL1/cLdMhzQ53eKwJ7TrH2xkznTVsq2nodnlWWaF6ZiEEEKIEfjXhpAYSYuHvswqGGJmwRetr+UbS2WAiMxZuElG9G1s0xeLzBBCCIkUVIgJiaG0eGYVDDHTsq31Jc31MiBLpKSyTkp2b2vTF4vMEEIIiRSWqu27ZMkSOfroo6V3796SmpoqeXl5MnHiRJkzZ06btj/88IMcdthhkpGRITk5OXLCCSfI+vXr/fb72GOPyZAhQyQ5OVn69esnt99+uzQ3N7dpV1ZWJjNmzJCCggJJS0tTx/70009Dcq4k9tFcGOCygNRmQEtxhqIeZqc40xcMOWpkkRw6tLtaYh0FQ7A/GMt2cW6qWmIdlm2j/fj2BbcNgKVvX2YekxBCCIlqhbiqqkp69eold999t3zwwQfy4osvSt++feW0006TO++809Nu1apVcsghh0hTU5O8/vrr8uyzz8ovv/wiBx10kOzYscOrz7vuuksuu+wypTB//PHHcuGFF6r+L7roIq92jY2NMmXKFKUAz5o1S/79739L9+7d5cgjj5TPP/88bDIgsYOW4gwpzeAmsXVnvVqGKsWZb8EQHAdLvUXaLMt2MGPqqC8zj0kIIYREtcsElFx89BxzzDGyYcMGmT17ttx8881q26233qqsve+9955kZbWG54wbN04GDRokDzzwgNx3331qW0VFhVKkzznnHKUEa8eAdRh9XX755TJs2DC1/ZlnnpHly5fLN998oyzDYPLkyTJ69Gi59tpr5dtvvw2rLEhsYHaKs44CzswqumFm8Q59X+6sZLXNX1+BHpOBd4QQQmJWIW4PuDDAnQE4nU6lCP/lL3/xKMOgT58+SoF95513PArxRx99JA0NDXLGGWd49Yf1m266SebOnetRiPG9wYMHe5RhYLfbZfr06XLjjTfK1q1bpbi4OExnTGIFs1KcGQk4M6vohpnFO/R9QdntnSVqmZGc6tWX0WMy8I4QQkjcKMQtLS3qs3PnTnnjjTeUq8Pjjz+u9q1bt07q6+tl1KhRbb6Hbf/73/+UEpySkqIsvmDkyJFe7YqKipSSre0H+DdcLvz1CVasWEGFOAqwquWwq2nxjAacmWWRNtOyrfW1bMM2EadI77w0GdmvqE1fRo7JwDtCCCGhwDoagw74+f79739X/3Y4HPLoo4/Keeed53GDAAi48wXb8JoVijSUXrSFa0V6errftlpfWr/t9ak/bnvAgu3rv7x27Vq1rKmpkV27dvn9Xm1trdeSBIezxS2L1lco62Jto1PSk+0qKGt8/3yxI/9YFMsceXuRsgxZGkZ3z2h1J8hKVkoxlMzR3R2SkmTztN9/r1QZ3b2H1DS6JCPZpvbV19ZIfYDHNasfra9BGQWy9MdNcsLIAsnLSfXbV0fHDFQOJDrne6xAuVPm8UKtBe8x0LtiQiGGi8LZZ5+tlMx3331XLr74YiXoq6++2tNGC7zxh36f0XaBtvXlySefVNkr/LFo0SIpLS3t8PtoQ7oO8twKdCKnSONmkS83x4bMcV5IWSaNvz10wf0A57ngq00STSz98fugvxtLcgg30TTfYwnKnTKPFxZZ6B5TUlISGwox0q7hA4466ii1vOGGG+T000+X/Pz8di22lZWVrZHpOTlqHW3hPlFXV6fSqPm2RSCeBtq21yfwZz32tWqffPLJbSzE06ZNk/Hjx8vQoUP9fg+KPiYR2vizZJPOgeUQxR6Q3xav2bWALOWzmpcm0yf08bIcRpvMAz0/q9JVuceKHMJNtM33WIFyp8zjhVoL3mNWrlwZGwqxLxDy008/rfIMQ4lFjuJly5a1aYdtAwcOVP7Det9hbN9///097WCtLS8vlxEjRni2oW17fQJ9W38UFhaqjz+QK1kfAOgPTKLO2hD/1FU3SHljotiT08Vpa732wJ7sVttb7KmSlZUStTLHCFHZDcUslm5v1AWcpSpf3ML8XIkmgpV7rMkh3ETLfI81KHfKPF5It9A9BnpXVOchbo958+ZJYmKi9O/fX2V+OPbYY+Xtt9+W3bt3e5nH0Q75hjWQQxjK8fPPP+/VH9ZhXYL1VuP4449X+Y316dWQ0QJFQaBM9+zZM+TnSaKjAEYkQMDZ0aN6Ksuow56ollgPNn1bPMgBAZal1Q0s6kEIIaRTLKUpnHvuuerpAhZhFMWAFRdZJl577TW55pprpFu31j968NXdb7/9VI7i66+/XrlFIDcxMkdcddVVnv7g5oB8w7fccov69xFHHCGLFy+WmTNnKh9lLeUaOPPMM+WJJ55Qbg/33nuvsvbCL3j16tXyySefREQeRMKeJizW07fFgxyYmo0QQkigWOovKnIAP/fcc/LCCy+oqnUweaMwxksvvaTyAWugDPP8+fPluuuuk5NOOklZjQ899FBVlENTmjWQbzgzM1Mpu9jfo0cPpURjux5ko0CVOhThuOSSS5Tf8ZgxY+TDDz+USZMmhU0GJLiUamYXwLAqXU3fpqe8pkF+rWpQhTAKMtq6lERrGjumZiOEEBIo1vkLt6dghm8RjfaAL7FRy+2ll16qPp0BqzSUcWItjFj8aEE1DpTcxz9bK1+vLfekqDtgYIFcfOjAqC+AgXPDmJGvGWn38JAEhR9vDvCwBMuylRR7Qggh1oB/GYjlCcTiZ6YFNVaBMvyfJVulwdki6Q6bbK2qV+uQ5zVTB0e1lRXWbCjwGLOWKlFlnkl1qDcH2M/5QQghxBfrmnoI8WPxK85NVUusw+KH/SQwNwlYhqEM98tPkx7ZqWqJ9a/W7FD7o1nm8RBgSQghxHyoEBNLY8TiR4wDn2HlJuGwqcwtAEusYzv2R7PMtQBLBFTCTWLrznq1jKUAS0IIIeZDhZhYmkha/MKdtiscx4M/LXyGa5tc0tLSorZhiXVsx/5ot7IyRR0hhJBAsfZfNhL3RCKlWrgDysJ5PGSTQAAdfIY3VNS1WoabXJJiT5QDB3XzZJuI5jR2DLAkhBASKNb+y0ZIBFKqhTugLNzHQzYJ9A2fYbhJFOc4lDJ80eQBMZXGjgGWhBBCjEKFmFiecFr8wp22KxJpwtAfskmccUCfdvMQ08pKCCEknqBCTKKGcFj8wp22K5JpwqAEt1eQQ4NWVkIIIfEAg+oI0RHugLJoD2AjhBBCYgH+tSUkgkF8kQgaJIQQQog3/GtLiA/hDiiLhQA2QgghJJqhQkyID+EOKGMAGyGEEBJZqBATYpGAMgawEUIIIZGBQXWEEEIIISSuoUJMCCGEEELiGirEhBBCCCEkrqFCTAghhBBC4hoqxISQqAMlr0urG9SSEEII6SrMMkEIiRqaXS0yb1WZfL9ppyp5jSp/KGyCXM5IX0cIIYQEAxViQkjUAGX4g2XbpKbRqar6rdtRo6r8obAJckcTQgghwUCTCiEkKoB7BCzDUIaHFWVJcW6qWmIdVf7oPkEIISRYqBATQqICVA2EmwQsw7AIAyyxjpLX2E8IIYQEAxViQkhUgBLa8Bmuqm8St9uttmGJ9cwUu9pPCCGEBAP/ghBCogKUtkYAHXyGf962S1mGoQxnJNtl3755YS2zTQghJLbgXxBCSNSAbBJwk4DPMNwkBnTLUMrwIYO7RXpohBBCohgqxISQqAGp1ZBN4oCB+cpnGG4StAwTQgjpKlSICSFRB5RgKsKEEELMgkF1hBBCCCEkrqFCTAghhBBC4hoqxIQQQgghJK6hQkwIIYQQQuIaKsSEEEIIISSuoUJMCCGEEELiGirEhBBCCCEkrqFCTAghhBBC4hoqxIQQQgghJK6hQkwIIYQQQuIalm4OIY2NjWq5du3adtvU1NRISUmJrFy5UjIyMkI5HEKZRxTOdco9nuB8p8zjhRoL6jGa3qXpYUagQhxCNm/erJbTpk0L5WEIIYQQQogfPWzs2LFihAS32+021JIETFVVlXz++efSq1cvSU5ObvcpBgrz3LlzZeDAgZRyGKDMIwPlTrnHE5zvlHm8sNaCegwsw1CGJ02aJDk5OYa+QwtxCMFFOO644wy1xSQaPnx4KIdDKHNLwLlOuccTnO+Uebww0GJ6jFHLsAaD6gghhBBCSFxDhZgQQgghhMQ1VIgJIYQQQkhcQ4U4wnTr1k1uu+02tSSUeSzDuU65xxOc75R5vNAtRvQYZpkghBBCCCFxDS3EhBBCCCEkrqFCTAghhBBC4hoqxIQQQgghJK6hQmwiu3fvlmuvvVaOOOII5VyekJAgM2fObNMO29v7DBkypE37xx57TG1Htbt+/frJ7bffLs3NzWYOPeZljoKM//jHP2TcuHGSlZUl+fn5qoLN+++/77dfytw8uT/66KOe+VtUVCQXXHCB7Ny5k3IPkM8++0zOPPNMJcv09HQpLi5WhX++//77Nm1/+OEHOeywwyQjI0MVCDrhhBNk/fr1lHkI5f7VV1/J2Wefre4xmOv4TWzcuLHdfnmP6brcXS6XPPTQQ3LkkUfKXnvtJWlpaTJ06FC5/vrrVaVYyj00c/3RRx+VCRMmSEFBgZrrvXv3llNOOUVWrFgR3TJH6WZiDhs2bHBnZ2e7Dz74YPfZZ5+Nktju2267rU27BQsWtPk88sgjqv3111/v1fbOO+90JyQkuG+44Qb3vHnz3H/729/cDofDfc455/CyBSDzW265Re07//zz3f/973/d//nPf9yHH3642vbWW29R5iGa61deeaU7MTHRfe211yq5Y55nZWW5x40b525qaqLcA+Ckk05yT5482f3kk0+658+f737jjTfcEyZMcNvtdvenn37qabdy5Up3Zmam+6CDDnK///77an4PHz7c3bNnT3dZWRllHiK5z5w5092nTx/3tGnT3Icccoj6TeB34g/e182R++7du9VcP/fcc9V+/I188MEH3bm5ue5hw4a56+rqKPcQzPVbb71Vzfd33nlHtXv22Wfde++9tzs9Pd29atWqqJU5FWITaWlpUR+wY8eOdpUEf8yYMUNNmjVr1ni2lZeXu1NSUtSPXc9dd92l2q5YscId7xiVeXFxsfvAAw/02lZfX6+Uuj/84Q+ebZS5eXLfsmWL22azuS+55BKv7a+88opqP3v2bMo9ALZv395mGxSC7t27u6dMmeLZdvLJJ7sLCgrc1dXVnm0bN250JyUlqQcTDc51c+Xucrk8/77//vvbVYgpd/Pk7nQ6lTx9gSIH+b/00kuUewjmuj9+/vlnJXMYn6J1rtNlwkQ0t4dgXj+/8cYb6hU+aoFrfPTRR9LQ0CBnnHGGV3us42Fm7ty5Eu8YlXlSUpJkZ2d7bUtJSfF8NChz8+S+cOFC9UrzqKOO8tp+zDHHqOVbb71FuQdAYWFhm21wiRg2bJhs3rxZrTudTnnvvffkxBNPVK5BGn369JHJkyfLO++8Q5mHQO4gMdHYn1PeY8yTu81mU+5vvowfP14t9deHcjdH5u2h5SC22+1RK3MqxBbgX//6l9TW1ir/Mz3Lly9Xy5EjR3pthx8mfHe0/aRzLrvsMvXjfOaZZ5T/6rZt2+TKK6+U6upqufTSSynzENDU1KSW8BvzfTiBMv3TTz9R7l0E8xf+wsOHD1fr69atk/r6ehk1alSbtti2du1a9QcK8P5intwDgXIPvdzhCwv07Sh382XucrmksbFRVq1apfQXKNR65TfaZP6bKk8iBpQ0BL7AqqOnoqJCKRNwbvclLy9P7SfGuPzyyyU1NVUuuugiz4MHZPjuu+/KAQccQJmHAFgVwNdff62skxrffPONsg7o5y/nenBgPuNh+qabbvLIUZvbvmAb5I4HQvxBoszNk3sgUO6hlfvWrVtVUN2+++7reRtFuYdG5unp6UohBnvvvbfMnz9fevXqFbUyp4U4wiAq89tvv5U///nPXq/uNTp6LR2Me0a88txzzykr8cUXXyyffPKJfPDBBypDAiJoP/74Y6+2lLk5jB49Wg4++GC5//77lUsQor6hDJ9//vnqVafvK2bKPTBuueUWefnll+Xhhx9WmQ2CkSVlbq7cjUK5h0bulZWVykULD36vvfYa7zEhlvk333wjCxYskDlz5khmZqYyfPhmmoimuU6F2ALWYeDrLgHgG4XXm3V1dX5/+P6sQKQtsIhpluEHHnhApkyZIr///e/l1Vdflf32208paJR5aIAiDAv8H//4R8nNzVU3TKQAGzNmjErpQ7kHB9IW3XnnnXLXXXephzy9HIE/ywvuGfgDhLdRWlveX8yReyBQ7qGRO+7zhx9+uLIQ/+9//5P+/ftT7iGW+dixY1X6NRj05s2bpx5Ebrzxxqid61SII+xj+dJLL6knLygIvmh+N8uWLfPaXlpaKuXl5TJixIiwjTWaWb16tfKrhPLrC16rIVdoTU2NWqfMzQU+ZbDGb9++XZYuXSplZWVyxx13yC+//KKsxxqUe2B/qJDzGR/9Hx8wYMAA5Rrke88A2IagXe1NFGVuntwDgXI3X+5QhpF3e8OGDUoZ9udDT7mbK3NfYCFGrmHc26NW5pFOcxGrGEm7pqWGQc4/f1RUVKiUJcidq+eee+6xZMoSq8p806ZNnhzEepA27IADDlA5K7UUYpS5eXJvj1mzZqncxN9//71nG+VujDvuuEPJ+uabb263zR//+Ed3YWGhe9euXV6/AeT+vO666yjzEMldT0dp1zjXzZV7ZWWle+zYse6cnBz34sWL221HuZsn8/b+DuBv6THHHBO1MmdQncl8+OGHygEdqdTAzz//LG+++ab6N3ybUElH7y4Ba86pp57qty+8Trj55puVHw/+DZ/XxYsXq6c2vP7Xgpbinc5kjio6eE0/e/Zs5eCPbQgEeOGFF1TA11//+lePLxNlbu5cR3VAzXIJH2J8B/P+7rvvVq/bNCj3znnwwQfl1ltvVVW5jj76aJXWTg9eXWrWHbwNQUARgovwyhLfQ1T3VVddRZmHSO47duyQzz//3MsihvmOdFT4IK0m57q5csebv6lTp8qPP/4ojzzyiEo7qG8HuePeQ7mbJ/Pq6mrlmgK9ZdCgQUqHgVV41qxZ6u/qbbfdFr339Uhr5LEGKhVBrP4+emtBSUmJspL95S9/MWRRQxUYWHh69+6tLHG+Vb7iGSMyRxEOWG1GjRqlKhvl5eWpCjxz5szxWIf1UObmyP3vf/+7e+jQoe60tDR3RkaGqp42d+7cdvuk3Ntn0qRJ7crb91b+3XffqUT6kDsqA6J62tq1aynzEModVbjaa4M+ONfNlzvuMx21Of300yl3k2Xe0NCgqpPivo57OqrY7bXXXu7p06e3a/GNlvt6Av4XaaWcEEIIIYSQSMGgOkIIIYQQEtdQISaEEEIIIXENFWJCCCGEEBLXUCEmhBBCCCFxDRViQgghhBAS11AhJoQQQgghcQ0VYkIIIYQQEtdQISaEEEIIIXENFWJCCCGEEBLXUCEmhMQczz//vCQkJPj9XH311Z52ffv2lRkzZnjWN27cqNrg+9HKIYccoj5W4cknn/Qrz/nz5ytZv/nmm13q/4477pBhw4ZJS0uLWIVbbrlFxo4da6kxEUI6xt7JfkIIiVqee+45GTJkiNe2nj17ttu+qKhIFixYIAMGDAjD6OIDKMQFBQVeDx5m8euvv8rf/vY3pXAnJlrHvoOHrscff1xeeOEFOeOMMyI9HEKIAagQE0JilhEjRsi+++5ruH1ycrJMmDBBrIzb7ZaGhgZJTU2VeGfWrFmSk5MjJ5xwgliJ7OxsmT59utx7773qQQCWcEKItbHOIzUhhEQYfy4TM2fOVNt+/PFHpXhlZWV5FJ4dO3Z4fR8uGMccc4y88847MmrUKElJSZH+/fvLo48+2uZYu3btUpbEfv36icPhkOLiYrn88sultrbWqx2OffHFF8vTTz8tQ4cOVUo7LI+B0NTUJHfeeaeyluP73bp1U5bL9sb/0UcfqVf+ULrxnWeffbZNn1999ZVMnDhRnSPGDjeBf/7zn2q8kKPW34oVK+Tzzz/3uKxgm57m5ma56aablOUesj3ssMNk9erVhs7pmWeekVNPPdXLOqxdw/vvv1/uu+8+dTycB9xIfvnlF3W866+/Xh0P1/H444+XsrIyv3J47733ZJ999lHfh+yxDjA/sJ6eni7jx4+X7777rs34TjvtNHW8efPmdXouhJDIQwsxISRmcblc4nQ6vbbZ7cHd9qA4/fGPf5Tzzz9fKXlQAH/++Wf59ttvJSkpydNuyZIlSrGFIt2jRw95+eWX5bLLLlMKnOa/XFdXJ5MmTZItW7bIjTfeqJRn9HnrrbfKsmXL5JNPPvGyKs6dO1e+/PJLtR99FhYWGh43/FiPO+449f1rr71Wfve738mmTZvktttuU0oilDm9tXnp0qVy1VVXKaWxe/fuSsk966yzZODAgXLwwQerNj/99JMcfvjhsvfeeyvlPC0tTSnsc+bM8To2HgxOOukkpXjCdQJAIdeD8z/ggAPUcfCQcN1118mxxx4rK1euFJvN1u55Qe4VFRUyefJkv/ufeOIJJVcsq6qq1Dmh3/33319dLyj5kAOuydlnny3/+c9/vL4POdxwww1KWcf4b7/9dvVAhG2ffvqp3H333eoaYbxQnjds2OAlx3HjxklGRoa8//77cuihhxq+XoSQCOEmhJAY47nnnnPj9ubv09zc7GnXp08f9+mnn+5Z37Bhg2qD72vcdtttatsVV1zhdYyXX35ZbZ8zZ45XfwkJCe4lS5Z4tT388MPdWVlZ7traWrV+zz33uBMTE92LFy/2avfmm2+qPj/44APPNqxnZ2e7KysrDZ37pEmT1Efj1VdfVX289dZbXu1wbGx/8sknvcafkpLi3rRpk2dbfX29Oy8vz33eeed5tp188snu9PR0944dOzzbXC6Xe9iwYapPyFFj+PDhXuPRmDdvnmp71FFHeW1//fXX1fYFCxZ0eJ733XefaldaWuq1XbuGo0ePVmPSeOSRR9T2P/zhD17tL7/8crW9urraSw6pqanuLVu2eLbhmqJdUVGR5zqCuXPnqu3/+c9/2ozxgAMOcO+///4dngchxBrQZYIQErO8+OKLsnjxYq9PsBbiP//5z17rsBajL99X4sOHD5fRo0d7bcNrfVg/f/jhB7WOV+/wbx4zZoyyYGufqVOnKqsjMjDogYUxNzc3qHHjWPCzhXVUfywcG9Zm32Nhe+/evT3rcImAJRjWVA24QGBMCJbTgNsCZBIof/jDH7zWYdUF+uO1F1AHWenHoOeoo47ycqWAiwM4+uijvdpp20tKStrIAa4gvu1gVYdF3He7v/HCkr9169YOz4MQYg3oMkEIiVmgrAQSVNcRUB71QBnOz89Xr+07aqffprXdvn27rF271svVQk95eXmb7BfBgmPBZQB+ykaOhXPyBW4O9fX1nnWcB9wpfPG3rTN8j6e5VOiP5w/sh/zac6vIy8vzWtfOv73tCFQ08/vaw0Rn50EIsQZUiAkhxAClpaVeFkNYWaEY+ip0aOfvu0BrC6sm/E39Batp+/V0JUsB+sJxESjnj8zMzID7RH9QtH3xd+6hAucFv2wEISK4zYpUVla2a8EmhFgLKsSEEGIABMchUErj9ddfV0qxbxEMBMchIEvvNvHKK68oxROZGwCCsBCUBcUSWSZCCY71r3/9SwUYIqDMDBAQ+MEHHyjrsqbwIXjvjTfe6NS6bBZaful169Z53Cysxvr165VrDCHE+lAhJoQQA7z99tvKTQLZFbQsE1B6ff1mkc4LfrHIMgFXB2Re+N///qdSgGm+p8hC8dZbb6msDVdccYVS6KBQwo/1v//9r8qIYJbyesoppyhlHj61yHaBNGFwNUCGC/g/IwMFMmgEAjIvvPvuuzJlyhT1b1i7kWVCSxmn990dOXKkUshfe+01lYIObgTY1lW0B5GFCxdaUiHG24M1a9bIJZdcEumhEEIMQIWYEEIMKsRQcp966inlwoAgtUceeaSNby6CsZDjF2nNoBBBQX7ooYeU4quBV/xIg4bCDbNnz/ak7EIwG/Lw+ubq7QrwsUVKMRSxeOmll+See+5Riv1ee+2lLL3BKKd4EICSj5Rlf/nLX1TAH/Luoj+kIUOaMg2kK9u2bZucc845snv3bunTp48nT3FX6NWrlxx00EHy73//W84991yxGhgXHjyCCTQkhISfBKSaiMBxCSEkKoASDKUORSw68weFIotX5FoBh3jjiCOOUMouClKEA1jZ/+///k9leND7d1sBKOt4wIF1nhBifWghJoQQEjBXXnmlquIGSy2Cx6D4wWqM6nHhAoUy9ttvP2X1fvzxx8UqfPHFFyrFX6AVBQkhkYMKMSGEkIBBkB4q5yGzBFxIhg0bplwyUNI6XOC4//jHP5RLCHyw9b7LkfYfRg5s+EwTQqIDukwQQgghhJC4xhqP04QQQgghhEQIKsSEEEIIISSuoUJMCCGEEELiGirEhBBCCCEkrqFCTAghhBBC4hoqxIQQQgghJK6hQkwIIYQQQuIaKsSEEEIIISSuoUJMCCGEEELiGirEhBBCCCEkrqFCTAghhBBCJJ75f+a5kM8BXYtnAAAAAElFTkSuQmCC", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } + "text/plain": [ + "shape: (1, 2)\n", + "┌─────────────────────┬───────────────────┐\n", + "│ corr_bill_len_depth ┆ corr_flipper_mass │\n", + "│ --- ┆ --- │\n", + "│ f64 ┆ f64 │\n", + "╞═════════════════════╪═══════════════════╡\n", + "│ -0.235053 ┆ 0.871202 │\n", + "└─────────────────────┴───────────────────┘" + ] + }, + "execution_count": 23, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "penguins_clean.select([\n", + " pl.corr('bill_length_mm', 'bill_depth_mm').alias('corr_bill_len_depth'),\n", + " pl.corr('flipper_length_mm', 'body_mass_g').alias('corr_flipper_mass'),\n", + "])" + ] + }, + { + "cell_type": "markdown", + "id": "589acfc7-b014-4553-ad65-5303ca727d22", + "metadata": {}, + "source": [ + "### Grouped Exploration\n", + "\n", + "Plotting bill depth against bill length for all penguins combined suggests that penguins with longer bill lengths have smaller depths. This observation matches the negative correlation between bill length and bill depth above.\n", + "- **Something suspicious?** What are those clusters of points and what do they represent?\n", + "- **Are we missing something?** Look for other variables that may be interfering with (confounding) the trend.\n", + "\n", + "Some simple techniques to involve a third variable in a bivariate plot are:\n", + "- **Color:** The most common technique used for both discrete and continuous variables.\n", + "- **Shape:** Common, but less used due to limitation in choices and distinguishability. Suitable for discrete variables only.\n", + "- **Size:** Less common, niche specific. Usually used for continuous variables.\n", + "- **Faceting** *(advanced)***:** Splits the plot into separate panels, one per group. Cleaner than overlaying for many or heavily overlapping groups. Suitable for discrete variables only.\n", + "\n", + "Let's color the points by **\"species\"** and re-examine the trend." + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "id": "8fafe78c-c1ab-47dd-a6ec-bae13b8baf7c", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "x = penguins_clean['bill_length_mm'].to_list()\n", + "y = penguins_clean['bill_depth_mm'].to_list()\n", + "\n", + "colors = {'Chinstrap': 'steelblue', 'Gentoo': 'salmon', 'Adelie': 'grey'}\n", + "species_col = penguins_clean['species'].to_list()\n", + "\n", + "plt.figure(figsize=(6,4))\n", + "for species, color in colors.items():\n", + " mask = [s == species for s in species_col]\n", + " plt.scatter([v for v, m in zip(x, mask) if m],\n", + " [v for v, m in zip(y, mask) if m],\n", + " s=12, alpha=0.6, color=color, label=species)\n", + "plt.xlabel('Bill length (mm)')\n", + "plt.ylabel('Bill depth (mm)')\n", + "plt.title('Penguins: bill length vs bill depth')\n", + "plt.legend(title='Species')\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "33409d63-47df-401e-985e-d172fbba61bc", + "metadata": {}, + "source": [ + "The relationship between bill length and bill depth now look positive within each species. This is a classic example of **Simpson's Paradox**. Let's now check out the correlations by species." + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "55f49121-dfb0-43f0-a943-0f7f2a566a3e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (3, 3)
speciescorr_bill_len_depthcorr_flipper_mass
strf64f64
"Chinstrap"0.6535360.641559
"Gentoo"0.6433840.702667
"Adelie"0.3914920.468202
" ], - "source": [ - "biscoe_mean_mass = (\n", - " penguins_clean\n", - " .filter(pl.col('island') == 'Biscoe')\n", - " .select(pl.col('body_mass_g').mean().alias('mean_body_mass_g'))\n", - ")\n", - "biscoe_mean_mass\n", - "\n", - "x = penguins_clean['flipper_length_mm'].to_list()\n", - "y = penguins_clean['body_mass_g'].to_list()\n", - "\n", - "plt.figure(figsize=(6,4))\n", - "plt.scatter(x, y, s=12, alpha=0.5)\n", - "plt.xlabel('Flipper length (mm)')\n", - "plt.ylabel('Body mass (g)')\n", - "plt.title('Penguins: flipper length vs body mass')\n", - "plt.tight_layout()\n", - "plt.show()" - ] - }, - { - "cell_type": "markdown", - "id": "dd7dc345", - "metadata": {}, - "source": [ - "## 7. Grouping + aggregation: choosing the grain\n", - "\n", - "**Grouping** decides what one row in your output represents.\n", - "\n", - "- `group_by(keys)` partitions rows into groups\n", - "- `.agg(...)` collapses each group (**information is lost**)\n", - "\n", - "```mermaid\n", - "flowchart LR\n", - " A[Row-level data] --> B[group_by keys]\n", - " B --> C[agg summaries]\n", - " C --> D[Fewer rows, more condensed meaning]\n", - "```\n", - "\n", - "Rule of thumb: do row-level feature creation first, then group/aggregate last.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 60, - "id": "ba38f7e4", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (10, 6)
CollegeWorkstudy Positionngpa_meangpa_medianrate_mean
strstru32f64f64f64
"Agriculture and Life Sciences""IT"52.7682.7316.054
"Agriculture and Life Sciences""Dining"23.213.2115.15
"Agriculture and Life Sciences""Campus Events"13.383.3814.52
"Agriculture and Life Sciences""Theater"13.323.3217.18
"Design""IT"23.6853.68516.78
"Design""Campus Events"23.2153.21518.35
"Design""Library"12.642.6416.97
"Design""Dining"12.422.4214.24
"Education""Library"33.5166673.5614.916667
"Education""Campus Events"33.3266673.416.713333
" - ], - "text/plain": [ - "shape: (10, 6)\n", - "┌───────────────────────────────┬────────────────────┬─────┬──────────┬────────────┬───────────┐\n", - "│ College ┆ Workstudy Position ┆ n ┆ gpa_mean ┆ gpa_median ┆ rate_mean │\n", - "│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", - "│ str ┆ str ┆ u32 ┆ f64 ┆ f64 ┆ f64 │\n", - "╞═══════════════════════════════╪════════════════════╪═════╪══════════╪════════════╪═══════════╡\n", - "│ Agriculture and Life Sciences ┆ IT ┆ 5 ┆ 2.768 ┆ 2.73 ┆ 16.054 │\n", - "│ Agriculture and Life Sciences ┆ Dining ┆ 2 ┆ 3.21 ┆ 3.21 ┆ 15.15 │\n", - "│ Agriculture and Life Sciences ┆ Campus Events ┆ 1 ┆ 3.38 ┆ 3.38 ┆ 14.52 │\n", - "│ Agriculture and Life Sciences ┆ Theater ┆ 1 ┆ 3.32 ┆ 3.32 ┆ 17.18 │\n", - "│ Design ┆ IT ┆ 2 ┆ 3.685 ┆ 3.685 ┆ 16.78 │\n", - "│ Design ┆ Campus Events ┆ 2 ┆ 3.215 ┆ 3.215 ┆ 18.35 │\n", - "│ Design ┆ Library ┆ 1 ┆ 2.64 ┆ 2.64 ┆ 16.97 │\n", - "│ Design ┆ Dining ┆ 1 ┆ 2.42 ┆ 2.42 ┆ 14.24 │\n", - "│ Education ┆ Library ┆ 3 ┆ 3.516667 ┆ 3.56 ┆ 14.916667 │\n", - "│ Education ┆ Campus Events ┆ 3 ┆ 3.326667 ┆ 3.4 ┆ 16.713333 │\n", - "└───────────────────────────────┴────────────────────┴─────┴──────────┴────────────┴───────────┘" - ] - }, - "execution_count": 60, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (3, 3)\n", + "┌───────────┬─────────────────────┬───────────────────┐\n", + "│ species ┆ corr_bill_len_depth ┆ corr_flipper_mass │\n", + "│ --- ┆ --- ┆ --- │\n", + "│ str ┆ f64 ┆ f64 │\n", + "╞═══════════╪═════════════════════╪═══════════════════╡\n", + "│ Chinstrap ┆ 0.653536 ┆ 0.641559 │\n", + "│ Gentoo ┆ 0.643384 ┆ 0.702667 │\n", + "│ Adelie ┆ 0.391492 ┆ 0.468202 │\n", + "└───────────┴─────────────────────┴───────────────────┘" + ] + }, + "execution_count": 25, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "penguins_clean.group_by('species').agg([\n", + " pl.corr('bill_length_mm', 'bill_depth_mm').alias('corr_bill_len_depth'),\n", + " pl.corr('flipper_length_mm', 'body_mass_g').alias('corr_flipper_mass'),\n", + "])" + ] + }, + { + "cell_type": "markdown", + "id": "37cea5f8-c546-4825-984e-9d967f5a764a", + "metadata": {}, + "source": [ + "The correlations corroborate our observations. Are there other patterns we are yet to uncover? Let's use different shapes to label the sexes." + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "id": "f166a52c-8399-429e-9472-95fad1444e48", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "markers = {'male': 'o', 'female': '^'}\n", + "sex_col = penguins_clean['sex'].to_list()\n", + "\n", + "plt.figure(figsize=(6,4))\n", + "for species, color in colors.items():\n", + " for sex, marker in markers.items():\n", + " mask = [s == species and g == sex for s, g in zip(species_col, sex_col)]\n", + " plt.scatter([v for v, m in zip(x, mask) if m],\n", + " [v for v, m in zip(y, mask) if m],\n", + " s=12, alpha=0.6, color=color, marker=marker, label=f'{species} / {sex}')\n", + "plt.xlabel('Bill length (mm)')\n", + "plt.ylabel('Bill depth (mm)')\n", + "plt.title('Penguins: bill length vs bill depth')\n", + "plt.legend(title = 'Species / Sex', loc = 'best')\n", + "plt.tight_layout()\n", + "plt.subplots_adjust(right=1.5, top=1.5)\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "f42393fb-e7ff-4a58-9f01-4692d3f06c41", + "metadata": {}, + "source": [ + "Within each species, females usually have shorter bill lengths and shallower bill depths than males." + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "id": "c50bcf16", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (3, 5)
speciesnmean_massmedian_massp90_mass
stru32f64f64f64
"Gentoo"1235076.016265000.05700.0
"Chinstrap"683733.0882353700.04150.0
"Adelie"1513700.6622523700.04300.0
" ], - "source": [ - "summary = (\n", - " celebs_typed\n", - " .group_by(['College', 'Workstudy Position'])\n", - " .agg([\n", - " pl.len().alias('n'),\n", - " pl.col('GPA').mean().alias('gpa_mean'),\n", - " pl.col('GPA').median().alias('gpa_median'),\n", - " pl.col('Workstudy Hourly Rate').mean().alias('rate_mean'),\n", - " ])\n", - " .sort(['College','n'], descending=[False, True])\n", - ")\n", - "summary.head(10)" - ] - }, - { - "cell_type": "markdown", - "id": "8ff0053b", - "metadata": {}, - "source": [ - "### Window functions: group-aware summaries without losing rows\n", - "\n", - "Sometimes you want group context **but you don’t want to collapse rows**.\n", - "\n", - "Window functions use `.over(group_keys)`.\n", - "\n", - "Mental model: **groupby + join back**, but simpler.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 61, - "id": "7ee9681a", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (8, 5)
NameCollegeGPAcollege_gpa_meangpa_minus_college_mean
strstrf64f64f64
"Adrian Wilson""Humanities and Social Sciences"3.883.5390.341
"Ahmet Özal""Education"3.543.4621430.077857
"Akihiro Kitamura""Natural Resources"2.172.91-0.74
"Anna Camp""Veterinary Medicine"2.182.677-0.497
"Anthony Mackie""Textiles"3.553.2490.301
"Archie Miller""Design"2.643.143333-0.503333
"Berlinda Tolbert""Agriculture and Life Sciences"3.272.9955560.274444
"Bill Cowher""Natural Resources"3.082.910.17
" - ], - "text/plain": [ - "shape: (8, 5)\n", - "┌──────────────────┬───────────────────────┬──────┬──────────────────┬────────────────────────┐\n", - "│ Name ┆ College ┆ GPA ┆ college_gpa_mean ┆ gpa_minus_college_mean │\n", - "│ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", - "│ str ┆ str ┆ f64 ┆ f64 ┆ f64 │\n", - "╞══════════════════╪═══════════════════════╪══════╪══════════════════╪════════════════════════╡\n", - "│ Adrian Wilson ┆ Humanities and Social ┆ 3.88 ┆ 3.539 ┆ 0.341 │\n", - "│ ┆ Sciences ┆ ┆ ┆ │\n", - "│ Ahmet Özal ┆ Education ┆ 3.54 ┆ 3.462143 ┆ 0.077857 │\n", - "│ Akihiro Kitamura ┆ Natural Resources ┆ 2.17 ┆ 2.91 ┆ -0.74 │\n", - "│ Anna Camp ┆ Veterinary Medicine ┆ 2.18 ┆ 2.677 ┆ -0.497 │\n", - "│ Anthony Mackie ┆ Textiles ┆ 3.55 ┆ 3.249 ┆ 0.301 │\n", - "│ Archie Miller ┆ Design ┆ 2.64 ┆ 3.143333 ┆ -0.503333 │\n", - "│ Berlinda Tolbert ┆ Agriculture and Life ┆ 3.27 ┆ 2.995556 ┆ 0.274444 │\n", - "│ ┆ Sciences ┆ ┆ ┆ │\n", - "│ Bill Cowher ┆ Natural Resources ┆ 3.08 ┆ 2.91 ┆ 0.17 │\n", - "└──────────────────┴───────────────────────┴──────┴──────────────────┴────────────────────────┘" - ] - }, - "execution_count": 61, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (3, 5)\n", + "┌───────────┬─────┬─────────────┬─────────────┬──────────┐\n", + "│ species ┆ n ┆ mean_mass ┆ median_mass ┆ p90_mass │\n", + "│ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", + "│ str ┆ u32 ┆ f64 ┆ f64 ┆ f64 │\n", + "╞═══════════╪═════╪═════════════╪═════════════╪══════════╡\n", + "│ Gentoo ┆ 123 ┆ 5076.01626 ┆ 5000.0 ┆ 5700.0 │\n", + "│ Chinstrap ┆ 68 ┆ 3733.088235 ┆ 3700.0 ┆ 4150.0 │\n", + "│ Adelie ┆ 151 ┆ 3700.662252 ┆ 3700.0 ┆ 4300.0 │\n", + "└───────────┴─────┴─────────────┴─────────────┴──────────┘" + ] + }, + "execution_count": 27, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "mass_by_species = (\n", + " penguins_clean\n", + " .group_by('species')\n", + " .agg([\n", + " pl.len().alias('n'),\n", + " pl.col('body_mass_g').mean().alias('mean_mass'),\n", + " pl.col('body_mass_g').median().alias('median_mass'),\n", + " pl.col('body_mass_g').quantile(0.9).alias('p90_mass'),\n", + " ])\n", + " .sort('mean_mass', descending=True)\n", + ")\n", + "mass_by_species" + ] + }, + { + "cell_type": "markdown", + "id": "20426f71", + "metadata": {}, + "source": [ + "### Try it yourself (bivariate)\n", + "\n", + "1) Filter penguins to a single island and compute mean body mass. \n", + "2) Make a scatter plot of flipper length vs body mass.\n", + "\n", + "- **Minimum**: island filter + mean body mass.\n", + "- **Stretch**: scatter plot; try `alpha=0.5` and a smaller `s` if it’s dense.\n", + "\n", + "
\n", + "Hint\n", + "Use `.filter(pl.col('island') == 'Biscoe')` and `plt.scatter(...)`.\n", + "
\n" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "id": "fc115819", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "biscoe_mean_mass = (\n", + " penguins_clean\n", + " .filter(pl.col('island') == 'Biscoe')\n", + " .select(pl.col('body_mass_g').mean().alias('mean_body_mass_g'))\n", + ")\n", + "biscoe_mean_mass\n", + "\n", + "x = penguins_clean['flipper_length_mm'].to_list()\n", + "y = penguins_clean['body_mass_g'].to_list()\n", + "\n", + "plt.figure(figsize=(6,4))\n", + "plt.scatter(x, y, s=12, alpha=0.5)\n", + "plt.xlabel('Flipper length (mm)')\n", + "plt.ylabel('Body mass (g)')\n", + "plt.title('Penguins: flipper length vs body mass')\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "dd7dc345", + "metadata": {}, + "source": [ + "## 7. Grouping + aggregation: choosing the grain\n", + "\n", + "**Grouping** decides what one row in your output represents.\n", + "\n", + "- `group_by(keys)` partitions rows into groups\n", + "- `.agg(...)` collapses each group (**information is lost**)\n", + "\n", + "```mermaid\n", + "flowchart LR\n", + " A[Row-level data] --> B[group_by keys]\n", + " B --> C[agg summaries]\n", + " C --> D[Fewer rows, more condensed meaning]\n", + "```\n", + "\n", + "Rule of thumb: do row-level feature creation first, then group/aggregate last.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "id": "ba38f7e4", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (10, 6)
CollegeWorkstudy Positionngpa_meangpa_medianrate_mean
strstru32f64f64f64
"Agriculture and Life Sciences""IT"52.7682.7316.054
"Agriculture and Life Sciences""Dining"23.213.2115.15
"Agriculture and Life Sciences""Theater"13.323.3217.18
"Agriculture and Life Sciences""Campus Events"13.383.3814.52
"Design""Campus Events"23.2153.21518.35
"Design""IT"23.6853.68516.78
"Design""Dining"12.422.4214.24
"Design""Library"12.642.6416.97
"Education""IT"33.7233333.6316.24
"Education""Library"33.5166673.5614.916667
" ], - "source": [ - "celebs_with_context = (\n", - " celebs_typed\n", - " .with_columns([\n", - " pl.col('GPA').mean().over('College').alias('college_gpa_mean'),\n", - " (pl.col('GPA') - pl.col('GPA').mean().over('College')).alias('gpa_minus_college_mean'),\n", - " ])\n", - ")\n", - "celebs_with_context.select(['Name','College','GPA','college_gpa_mean','gpa_minus_college_mean']).head(8)" - ] - }, - { - "cell_type": "markdown", - "id": "98cd06ec", - "metadata": {}, - "source": [ - "## 8. (OPTIONAL) Pivoting = groupby + aggregation + reshaping\n", - "\n", - "A pivot table answers: **“Which values should become columns?”**\n", - "\n", - "A pivot needs *one value per cell*. If multiple rows map to the same cell, you must choose an aggregation.\n", - "\n", - "Here we pivot average GPA by College (rows) and Workstudy Position (columns).\n" - ] - }, - { - "cell_type": "code", - "execution_count": 62, - "id": "459a0678", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (11, 7)
CollegeITCampus EventsLibraryDiningAV AssistantTheater
strf64f64f64f64f64f64
"Humanities and Social Sciences"3.6752.793.753.42null3.91
"Education"3.7233333.3266673.5166673.53.163.305
"Natural Resources"2.78252.473.613.19null2.475
"Veterinary Medicine"2.3652.332.7252.8933333.25null
"Textiles"2.773.293.553.6366672.88null
…………………
"Agriculture and Life Sciences"2.7683.38null3.21null3.32
"Poole College of Management"2.903333null3.743.453.722.88
"Engineering"3.062.743.193.406667nullnull
"Sciences"2.17null3.1833332.7833334.254.58
"University College"nullnull3.1566673.49nullnull
" - ], - "text/plain": [ - "shape: (11, 7)\n", - "┌────────────────────────┬──────────┬───────────────┬──────────┬──────────┬──────────────┬─────────┐\n", - "│ College ┆ IT ┆ Campus Events ┆ Library ┆ Dining ┆ AV Assistant ┆ Theater │\n", - "│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", - "│ str ┆ f64 ┆ f64 ┆ f64 ┆ f64 ┆ f64 ┆ f64 │\n", - "╞════════════════════════╪══════════╪═══════════════╪══════════╪══════════╪══════════════╪═════════╡\n", - "│ Humanities and Social ┆ 3.675 ┆ 2.79 ┆ 3.75 ┆ 3.42 ┆ null ┆ 3.91 │\n", - "│ Sciences ┆ ┆ ┆ ┆ ┆ ┆ │\n", - "│ Education ┆ 3.723333 ┆ 3.326667 ┆ 3.516667 ┆ 3.5 ┆ 3.16 ┆ 3.305 │\n", - "│ Natural Resources ┆ 2.7825 ┆ 2.47 ┆ 3.61 ┆ 3.19 ┆ null ┆ 2.475 │\n", - "│ Veterinary Medicine ┆ 2.365 ┆ 2.33 ┆ 2.725 ┆ 2.893333 ┆ 3.25 ┆ null │\n", - "│ Textiles ┆ 2.77 ┆ 3.29 ┆ 3.55 ┆ 3.636667 ┆ 2.88 ┆ null │\n", - "│ … ┆ … ┆ … ┆ … ┆ … ┆ … ┆ … │\n", - "│ Agriculture and Life ┆ 2.768 ┆ 3.38 ┆ null ┆ 3.21 ┆ null ┆ 3.32 │\n", - "│ Sciences ┆ ┆ ┆ ┆ ┆ ┆ │\n", - "│ Poole College of ┆ 2.903333 ┆ null ┆ 3.74 ┆ 3.45 ┆ 3.72 ┆ 2.88 │\n", - "│ Management ┆ ┆ ┆ ┆ ┆ ┆ │\n", - "│ Engineering ┆ 3.06 ┆ 2.74 ┆ 3.19 ┆ 3.406667 ┆ null ┆ null │\n", - "│ Sciences ┆ 2.17 ┆ null ┆ 3.183333 ┆ 2.783333 ┆ 4.25 ┆ 4.58 │\n", - "│ University College ┆ null ┆ null ┆ 3.156667 ┆ 3.49 ┆ null ┆ null │\n", - "└────────────────────────┴──────────┴───────────────┴──────────┴──────────┴──────────────┴─────────┘" - ] - }, - "execution_count": 62, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (10, 6)\n", + "┌───────────────────────────────┬────────────────────┬─────┬──────────┬────────────┬───────────┐\n", + "│ College ┆ Workstudy Position ┆ n ┆ gpa_mean ┆ gpa_median ┆ rate_mean │\n", + "│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", + "│ str ┆ str ┆ u32 ┆ f64 ┆ f64 ┆ f64 │\n", + "╞═══════════════════════════════╪════════════════════╪═════╪══════════╪════════════╪═══════════╡\n", + "│ Agriculture and Life Sciences ┆ IT ┆ 5 ┆ 2.768 ┆ 2.73 ┆ 16.054 │\n", + "│ Agriculture and Life Sciences ┆ Dining ┆ 2 ┆ 3.21 ┆ 3.21 ┆ 15.15 │\n", + "│ Agriculture and Life Sciences ┆ Theater ┆ 1 ┆ 3.32 ┆ 3.32 ┆ 17.18 │\n", + "│ Agriculture and Life Sciences ┆ Campus Events ┆ 1 ┆ 3.38 ┆ 3.38 ┆ 14.52 │\n", + "│ Design ┆ Campus Events ┆ 2 ┆ 3.215 ┆ 3.215 ┆ 18.35 │\n", + "│ Design ┆ IT ┆ 2 ┆ 3.685 ┆ 3.685 ┆ 16.78 │\n", + "│ Design ┆ Dining ┆ 1 ┆ 2.42 ┆ 2.42 ┆ 14.24 │\n", + "│ Design ┆ Library ┆ 1 ┆ 2.64 ┆ 2.64 ┆ 16.97 │\n", + "│ Education ┆ IT ┆ 3 ┆ 3.723333 ┆ 3.63 ┆ 16.24 │\n", + "│ Education ┆ Library ┆ 3 ┆ 3.516667 ┆ 3.56 ┆ 14.916667 │\n", + "└───────────────────────────────┴────────────────────┴─────┴──────────┴────────────┴───────────┘" + ] + }, + "execution_count": 29, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "summary = (\n", + " celebs_typed\n", + " .group_by(['College', 'Workstudy Position'])\n", + " .agg([\n", + " pl.len().alias('n'),\n", + " pl.col('GPA').mean().alias('gpa_mean'),\n", + " pl.col('GPA').median().alias('gpa_median'),\n", + " pl.col('Workstudy Hourly Rate').mean().alias('rate_mean'),\n", + " ])\n", + " .sort(['College','n'], descending=[False, True])\n", + ")\n", + "summary.head(10)" + ] + }, + { + "cell_type": "markdown", + "id": "8ff0053b", + "metadata": {}, + "source": [ + "### Window functions: group-aware summaries without losing rows\n", + "\n", + "Sometimes you want group context **but you don’t want to collapse rows**.\n", + "\n", + "Window functions use `.over(group_keys)`.\n", + "\n", + "Mental model: **groupby + join back**, but simpler.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "id": "7ee9681a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (8, 5)
NameCollegeGPAcollege_gpa_meangpa_minus_college_mean
strstrf64f64f64
"Adrian Wilson""Humanities and Social Sciences"3.883.5390.341
"Ahmet Özal""Education"3.543.4621430.077857
"Akihiro Kitamura""Natural Resources"2.172.91-0.74
"Anna Camp""Veterinary Medicine"2.182.677-0.497
"Anthony Mackie""Textiles"3.553.2490.301
"Archie Miller""Design"2.643.143333-0.503333
"Berlinda Tolbert""Agriculture and Life Sciences"3.272.9955560.274444
"Bill Cowher""Natural Resources"3.082.910.17
" ], - "source": [ - "pivot_gpa = celebs_typed.pivot(\n", - " values='GPA',\n", - " index='College',\n", - " on='Workstudy Position',\n", - " aggregate_function='mean'\n", - ")\n", - "pivot_gpa" - ] - }, - { - "cell_type": "markdown", - "id": "6f56dfe4", - "metadata": {}, - "source": [ - "### (OPTIONAL) Try it yourself (pivot)\n", - "\n", - "Create a pivot of **mean hourly rate** by `College` (rows) and `Workstudy Position` (columns).\n", - "\n", - "
\n", - "Solution\n", - "\n", - "```python\n", - "celebs_typed.pivot(\n", - " values='Workstudy Hourly Rate',\n", - " index='College',\n", - " on='Workstudy Position',\n", - " aggregate_function='mean'\n", - ")\n", - "```\n", - "\n", - "
\n" - ] - }, - { - "cell_type": "code", - "execution_count": 63, - "id": "233ac59e", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (11, 7)
CollegeITCampus EventsLibraryDiningAV AssistantTheater
strf64f64f64f64f64f64
"Humanities and Social Sciences"16.0417.2415.7816.5125null14.72
"Education"16.2416.71333314.91666713.97516.8717.84
"Natural Resources"17.092514.7617.0116.255null16.665
"Veterinary Medicine"16.43515.35514.93515.3916.96null
"Textiles"16.11666716.319.2916.50666716.1null
…………………
"Agriculture and Life Sciences"16.05414.52null15.15null17.18
"Poole College of Management"15.826667null16.5417.7515.9914.925
"Engineering"16.5216.0516.4817.283333nullnull
"Sciences"16.535null16.88666715.56666716.6813.79
"University College"nullnull16.43666715.47nullnull
" - ], - "text/plain": [ - "shape: (11, 7)\n", - "┌─────────────────────┬───────────┬───────────────┬───────────┬───────────┬──────────────┬─────────┐\n", - "│ College ┆ IT ┆ Campus Events ┆ Library ┆ Dining ┆ AV Assistant ┆ Theater │\n", - "│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", - "│ str ┆ f64 ┆ f64 ┆ f64 ┆ f64 ┆ f64 ┆ f64 │\n", - "╞═════════════════════╪═══════════╪═══════════════╪═══════════╪═══════════╪══════════════╪═════════╡\n", - "│ Humanities and ┆ 16.04 ┆ 17.24 ┆ 15.78 ┆ 16.5125 ┆ null ┆ 14.72 │\n", - "│ Social Sciences ┆ ┆ ┆ ┆ ┆ ┆ │\n", - "│ Education ┆ 16.24 ┆ 16.713333 ┆ 14.916667 ┆ 13.975 ┆ 16.87 ┆ 17.84 │\n", - "│ Natural Resources ┆ 17.0925 ┆ 14.76 ┆ 17.01 ┆ 16.255 ┆ null ┆ 16.665 │\n", - "│ Veterinary Medicine ┆ 16.435 ┆ 15.355 ┆ 14.935 ┆ 15.39 ┆ 16.96 ┆ null │\n", - "│ Textiles ┆ 16.116667 ┆ 16.3 ┆ 19.29 ┆ 16.506667 ┆ 16.1 ┆ null │\n", - "│ … ┆ … ┆ … ┆ … ┆ … ┆ … ┆ … │\n", - "│ Agriculture and ┆ 16.054 ┆ 14.52 ┆ null ┆ 15.15 ┆ null ┆ 17.18 │\n", - "│ Life Sciences ┆ ┆ ┆ ┆ ┆ ┆ │\n", - "│ Poole College of ┆ 15.826667 ┆ null ┆ 16.54 ┆ 17.75 ┆ 15.99 ┆ 14.925 │\n", - "│ Management ┆ ┆ ┆ ┆ ┆ ┆ │\n", - "│ Engineering ┆ 16.52 ┆ 16.05 ┆ 16.48 ┆ 17.283333 ┆ null ┆ null │\n", - "│ Sciences ┆ 16.535 ┆ null ┆ 16.886667 ┆ 15.566667 ┆ 16.68 ┆ 13.79 │\n", - "│ University College ┆ null ┆ null ┆ 16.436667 ┆ 15.47 ┆ null ┆ null │\n", - "└─────────────────────┴───────────┴───────────────┴───────────┴───────────┴──────────────┴─────────┘" - ] - }, - "execution_count": 63, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (8, 5)\n", + "┌──────────────────┬───────────────────────┬──────┬──────────────────┬────────────────────────┐\n", + "│ Name ┆ College ┆ GPA ┆ college_gpa_mean ┆ gpa_minus_college_mean │\n", + "│ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", + "│ str ┆ str ┆ f64 ┆ f64 ┆ f64 │\n", + "╞══════════════════╪═══════════════════════╪══════╪══════════════════╪════════════════════════╡\n", + "│ Adrian Wilson ┆ Humanities and Social ┆ 3.88 ┆ 3.539 ┆ 0.341 │\n", + "│ ┆ Sciences ┆ ┆ ┆ │\n", + "│ Ahmet Özal ┆ Education ┆ 3.54 ┆ 3.462143 ┆ 0.077857 │\n", + "│ Akihiro Kitamura ┆ Natural Resources ┆ 2.17 ┆ 2.91 ┆ -0.74 │\n", + "│ Anna Camp ┆ Veterinary Medicine ┆ 2.18 ┆ 2.677 ┆ -0.497 │\n", + "│ Anthony Mackie ┆ Textiles ┆ 3.55 ┆ 3.249 ┆ 0.301 │\n", + "│ Archie Miller ┆ Design ┆ 2.64 ┆ 3.143333 ┆ -0.503333 │\n", + "│ Berlinda Tolbert ┆ Agriculture and Life ┆ 3.27 ┆ 2.995556 ┆ 0.274444 │\n", + "│ ┆ Sciences ┆ ┆ ┆ │\n", + "│ Bill Cowher ┆ Natural Resources ┆ 3.08 ┆ 2.91 ┆ 0.17 │\n", + "└──────────────────┴───────────────────────┴──────┴──────────────────┴────────────────────────┘" + ] + }, + "execution_count": 30, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "celebs_with_context = (\n", + " celebs_typed\n", + " .with_columns([\n", + " pl.col('GPA').mean().over('College').alias('college_gpa_mean'),\n", + " (pl.col('GPA') - pl.col('GPA').mean().over('College')).alias('gpa_minus_college_mean'),\n", + " ])\n", + ")\n", + "celebs_with_context.select(['Name','College','GPA','college_gpa_mean','gpa_minus_college_mean']).head(8)" + ] + }, + { + "cell_type": "markdown", + "id": "98cd06ec", + "metadata": {}, + "source": [ + "## 8. (OPTIONAL) Pivoting = groupby + aggregation + reshaping\n", + "\n", + "A pivot table answers: **“Which values should become columns?”**\n", + "\n", + "A pivot needs *one value per cell*. If multiple rows map to the same cell, you must choose an aggregation.\n", + "\n", + "Here we pivot average GPA by College (rows) and Workstudy Position (columns).\n" + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "id": "459a0678", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (11, 7)
CollegeITCampus EventsLibraryDiningAV AssistantTheater
strf64f64f64f64f64f64
"Humanities and Social Sciences"3.6752.793.753.42null3.91
"Education"3.7233333.3266673.5166673.53.163.305
"Natural Resources"2.78252.473.613.19null2.475
"Veterinary Medicine"2.3652.332.7252.8933333.25null
"Textiles"2.773.293.553.6366672.88null
…………………
"Agriculture and Life Sciences"2.7683.38null3.21null3.32
"Poole College of Management"2.903333null3.743.453.722.88
"Engineering"3.062.743.193.406667nullnull
"Sciences"2.17null3.1833332.7833334.254.58
"University College"nullnull3.1566673.49nullnull
" ], - "source": [ - "celebs_typed.pivot(\n", - " values='Workstudy Hourly Rate',\n", - " index='College',\n", - " on='Workstudy Position',\n", - " aggregate_function='mean'\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "1dd671ed", - "metadata": {}, - "source": [ - "## 9. (OPTIONAL) Joins (combining tables)\n", - "\n", - "Joins combine columns from two tables using keys.\n", - "\n", - "In real projects, joins often introduce missingness (unmatched keys) — so always inspect results.\n", - "\n", - "Below is a tiny, self-contained example (not tied to the workshop datasets).\n" - ] - }, - { - "cell_type": "code", - "execution_count": 64, - "id": "7b5a5c1e", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (3, 3)
idxy
i64i64str
110null
220"a"
330"b"
" - ], - "text/plain": [ - "shape: (3, 3)\n", - "┌─────┬─────┬──────┐\n", - "│ id ┆ x ┆ y │\n", - "│ --- ┆ --- ┆ --- │\n", - "│ i64 ┆ i64 ┆ str │\n", - "╞═════╪═════╪══════╡\n", - "│ 1 ┆ 10 ┆ null │\n", - "│ 2 ┆ 20 ┆ a │\n", - "│ 3 ┆ 30 ┆ b │\n", - "└─────┴─────┴──────┘" - ] - }, - "execution_count": 64, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (11, 7)\n", + "┌────────────────────────┬──────────┬───────────────┬──────────┬──────────┬──────────────┬─────────┐\n", + "│ College ┆ IT ┆ Campus Events ┆ Library ┆ Dining ┆ AV Assistant ┆ Theater │\n", + "│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", + "│ str ┆ f64 ┆ f64 ┆ f64 ┆ f64 ┆ f64 ┆ f64 │\n", + "╞════════════════════════╪══════════╪═══════════════╪══════════╪══════════╪══════════════╪═════════╡\n", + "│ Humanities and Social ┆ 3.675 ┆ 2.79 ┆ 3.75 ┆ 3.42 ┆ null ┆ 3.91 │\n", + "│ Sciences ┆ ┆ ┆ ┆ ┆ ┆ │\n", + "│ Education ┆ 3.723333 ┆ 3.326667 ┆ 3.516667 ┆ 3.5 ┆ 3.16 ┆ 3.305 │\n", + "│ Natural Resources ┆ 2.7825 ┆ 2.47 ┆ 3.61 ┆ 3.19 ┆ null ┆ 2.475 │\n", + "│ Veterinary Medicine ┆ 2.365 ┆ 2.33 ┆ 2.725 ┆ 2.893333 ┆ 3.25 ┆ null │\n", + "│ Textiles ┆ 2.77 ┆ 3.29 ┆ 3.55 ┆ 3.636667 ┆ 2.88 ┆ null │\n", + "│ … ┆ … ┆ … ┆ … ┆ … ┆ … ┆ … │\n", + "│ Agriculture and Life ┆ 2.768 ┆ 3.38 ┆ null ┆ 3.21 ┆ null ┆ 3.32 │\n", + "│ Sciences ┆ ┆ ┆ ┆ ┆ ┆ │\n", + "│ Poole College of ┆ 2.903333 ┆ null ┆ 3.74 ┆ 3.45 ┆ 3.72 ┆ 2.88 │\n", + "│ Management ┆ ┆ ┆ ┆ ┆ ┆ │\n", + "│ Engineering ┆ 3.06 ┆ 2.74 ┆ 3.19 ┆ 3.406667 ┆ null ┆ null │\n", + "│ Sciences ┆ 2.17 ┆ null ┆ 3.183333 ┆ 2.783333 ┆ 4.25 ┆ 4.58 │\n", + "│ University College ┆ null ┆ null ┆ 3.156667 ┆ 3.49 ┆ null ┆ null │\n", + "└────────────────────────┴──────────┴───────────────┴──────────┴──────────┴──────────────┴─────────┘" + ] + }, + "execution_count": 31, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "pivot_gpa = celebs_typed.pivot(\n", + " values='GPA',\n", + " index='College',\n", + " on='Workstudy Position',\n", + " aggregate_function='mean'\n", + ")\n", + "pivot_gpa" + ] + }, + { + "cell_type": "markdown", + "id": "6f56dfe4", + "metadata": {}, + "source": [ + "### (OPTIONAL) Try it yourself (pivot)\n", + "\n", + "Create a pivot of **mean hourly rate** by `College` (rows) and `Workstudy Position` (columns).\n", + "\n", + "
\n", + "Solution\n", + "\n", + "```python\n", + "celebs_typed.pivot(\n", + " values='Workstudy Hourly Rate',\n", + " index='College',\n", + " on='Workstudy Position',\n", + " aggregate_function='mean'\n", + ")\n", + "```\n", + "\n", + "
\n" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "id": "233ac59e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (11, 7)
CollegeITCampus EventsLibraryDiningAV AssistantTheater
strf64f64f64f64f64f64
"Humanities and Social Sciences"16.0417.2415.7816.5125null14.72
"Education"16.2416.71333314.91666713.97516.8717.84
"Natural Resources"17.092514.7617.0116.255null16.665
"Veterinary Medicine"16.43515.35514.93515.3916.96null
"Textiles"16.11666716.319.2916.50666716.1null
…………………
"Agriculture and Life Sciences"16.05414.52null15.15null17.18
"Poole College of Management"15.826667null16.5417.7515.9914.925
"Engineering"16.5216.0516.4817.283333nullnull
"Sciences"16.535null16.88666715.56666716.6813.79
"University College"nullnull16.43666715.47nullnull
" ], - "source": [ - "left = pl.DataFrame({'id':[1,2,3], 'x':[10,20,30]})\n", - "right = pl.DataFrame({'id':[2,3,4], 'y':['a','b','c']})\n", - "\n", - "left.join(right, on='id', how='left')" - ] - }, - { - "cell_type": "markdown", - "id": "955433e8", - "metadata": {}, - "source": [ - "## 10. (OPTIONAL) Lazy execution + query plans\n", - "\n", - "Use lazy mode when:\n", - "- files are large\n", - "- you want to push filters into the scan\n", - "- you want Polars to optimize the whole query\n", - "\n", - "```mermaid\n", - "flowchart LR\n", - " A[scan_*] --> B[build lazy plan]\n", - " B --> C[explain]\n", - " C --> D[collect]\n", - " D --> E[result DataFrame]\n", - "```\n", - "\n", - "Key idea: `scan_*` builds a plan; `collect()` executes.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 65, - "id": "4ca151c5", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "naive plan: (run LazyFrame.explain(optimized=True) to see the optimized plan)\n", - "

\n", - "
SORT BY [descending: [true]] [col(\"n\")]

AGGREGATE[maintain_order: false]

[len().alias(\"n\")] BY [col(\"Species\")]

FROM

FILTER col(\"Species\").is_not_null()

FROM

Csv SCAN [data/NCSU_Mascots_v1.csv]

PROJECT */22 COLUMNS

ESTIMATED ROWS: 59
" - ], - "text/plain": [ - "" - ] - }, - "execution_count": 65, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (11, 7)\n", + "┌─────────────────────┬───────────┬───────────────┬───────────┬───────────┬──────────────┬─────────┐\n", + "│ College ┆ IT ┆ Campus Events ┆ Library ┆ Dining ┆ AV Assistant ┆ Theater │\n", + "│ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- ┆ --- │\n", + "│ str ┆ f64 ┆ f64 ┆ f64 ┆ f64 ┆ f64 ┆ f64 │\n", + "╞═════════════════════╪═══════════╪═══════════════╪═══════════╪═══════════╪══════════════╪═════════╡\n", + "│ Humanities and ┆ 16.04 ┆ 17.24 ┆ 15.78 ┆ 16.5125 ┆ null ┆ 14.72 │\n", + "│ Social Sciences ┆ ┆ ┆ ┆ ┆ ┆ │\n", + "│ Education ┆ 16.24 ┆ 16.713333 ┆ 14.916667 ┆ 13.975 ┆ 16.87 ┆ 17.84 │\n", + "│ Natural Resources ┆ 17.0925 ┆ 14.76 ┆ 17.01 ┆ 16.255 ┆ null ┆ 16.665 │\n", + "│ Veterinary Medicine ┆ 16.435 ┆ 15.355 ┆ 14.935 ┆ 15.39 ┆ 16.96 ┆ null │\n", + "│ Textiles ┆ 16.116667 ┆ 16.3 ┆ 19.29 ┆ 16.506667 ┆ 16.1 ┆ null │\n", + "│ … ┆ … ┆ … ┆ … ┆ … ┆ … ┆ … │\n", + "│ Agriculture and ┆ 16.054 ┆ 14.52 ┆ null ┆ 15.15 ┆ null ┆ 17.18 │\n", + "│ Life Sciences ┆ ┆ ┆ ┆ ┆ ┆ │\n", + "│ Poole College of ┆ 15.826667 ┆ null ┆ 16.54 ┆ 17.75 ┆ 15.99 ┆ 14.925 │\n", + "│ Management ┆ ┆ ┆ ┆ ┆ ┆ │\n", + "│ Engineering ┆ 16.52 ┆ 16.05 ┆ 16.48 ┆ 17.283333 ┆ null ┆ null │\n", + "│ Sciences ┆ 16.535 ┆ null ┆ 16.886667 ┆ 15.566667 ┆ 16.68 ┆ 13.79 │\n", + "│ University College ┆ null ┆ null ┆ 16.436667 ┆ 15.47 ┆ null ┆ null │\n", + "└─────────────────────┴───────────┴───────────────┴───────────┴───────────┴──────────────┴─────────┘" + ] + }, + "execution_count": 32, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "celebs_typed.pivot(\n", + " values='Workstudy Hourly Rate',\n", + " index='College',\n", + " on='Workstudy Position',\n", + " aggregate_function='mean'\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "1dd671ed", + "metadata": {}, + "source": [ + "## 9. (OPTIONAL) Joins (combining tables)\n", + "\n", + "Joins combine columns from two tables using keys.\n", + "\n", + "In real projects, joins often introduce missingness (unmatched keys) — so always inspect results.\n", + "\n", + "Below is a tiny, self-contained example (not tied to the workshop datasets).\n" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "id": "7b5a5c1e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (3, 3)
idxy
i64i64str
110null
220"a"
330"b"
" ], - "source": [ - "lazy_plan = (\n", - " pl.scan_csv(data_dir / 'NCSU_Mascots_v1.csv')\n", - " .filter(pl.col('Species').is_not_null())\n", - " .group_by('Species')\n", - " .agg(pl.len().alias('n'))\n", - " .sort('n', descending=True)\n", - ")\n", - "\n", - "lazy_plan # shows a LazyFrame (a plan)" - ] - }, - { - "cell_type": "code", - "execution_count": 66, - "id": "4d489f20", - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "'SORT BY [descending: [true]] [col(\"n\")]\\n AGGREGATE[maintain_order: false]\\n [len().alias(\"n\")] BY [col(\"Species\")]\\n FROM\\n Csv SCAN [data/NCSU_Mascots_v1.csv]\\n PROJECT 1/22 COLUMNS\\n SELECTION: col(\"Species\").is_not_null()\\n ESTIMATED ROWS: 59'" - ] - }, - "execution_count": 66, - "metadata": {}, - "output_type": "execute_result" - } + "text/plain": [ + "shape: (3, 3)\n", + "┌─────┬─────┬──────┐\n", + "│ id ┆ x ┆ y │\n", + "│ --- ┆ --- ┆ --- │\n", + "│ i64 ┆ i64 ┆ str │\n", + "╞═════╪═════╪══════╡\n", + "│ 1 ┆ 10 ┆ null │\n", + "│ 2 ┆ 20 ┆ a │\n", + "│ 3 ┆ 30 ┆ b │\n", + "└─────┴─────┴──────┘" + ] + }, + "execution_count": 33, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "left = pl.DataFrame({'id':[1,2,3], 'x':[10,20,30]})\n", + "right = pl.DataFrame({'id':[2,3,4], 'y':['a','b','c']})\n", + "\n", + "left.join(right, on='id', how='left')" + ] + }, + { + "cell_type": "markdown", + "id": "955433e8", + "metadata": {}, + "source": [ + "## 10. (OPTIONAL) Lazy execution + query plans\n", + "\n", + "Use lazy mode when:\n", + "- files are large\n", + "- you want to push filters into the scan\n", + "- you want Polars to optimize the whole query\n", + "\n", + "```mermaid\n", + "flowchart LR\n", + " A[scan_*] --> B[build lazy plan]\n", + " B --> C[explain]\n", + " C --> D[collect]\n", + " D --> E[result DataFrame]\n", + "```\n", + "\n", + "Key idea: `scan_*` builds a plan; `collect()` executes.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "id": "4ca151c5", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "naive plan: (run LazyFrame.explain(optimized=True) to see the optimized plan)\n", + "

\n", + "
SORT BY [descending: [true]] [col(\"n\")]

AGGREGATE[maintain_order: false]

[len().alias(\"n\")] BY [col(\"Species\")]

FROM

FILTER col(\"Species\").is_not_null()

FROM

Csv SCAN [data/NCSU_Mascots_v1.csv]

PROJECT */22 COLUMNS

ESTIMATED ROWS: 59
" ], - "source": [ - "lazy_plan.explain()" + "text/plain": [ + "" ] - }, - { - "cell_type": "code", - "execution_count": 67, - "id": "a18abc8f", - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "shape: (10, 2)
Speciesn
stru32
"Dog"10
"Cat"5
"Bird"4
"Rabbit"3
"Wolf"3
"Horse"3
"Lizard"3
"Turtle"2
"Fish"2
"Hamster"2
" - ], - "text/plain": [ - "shape: (10, 2)\n", - "┌─────────┬─────┐\n", - "│ Species ┆ n │\n", - "│ --- ┆ --- │\n", - "│ str ┆ u32 │\n", - "╞═════════╪═════╡\n", - "│ Dog ┆ 10 │\n", - "│ Cat ┆ 5 │\n", - "│ Bird ┆ 4 │\n", - "│ Rabbit ┆ 3 │\n", - "│ Wolf ┆ 3 │\n", - "│ Horse ┆ 3 │\n", - "│ Lizard ┆ 3 │\n", - "│ Turtle ┆ 2 │\n", - "│ Fish ┆ 2 │\n", - "│ Hamster ┆ 2 │\n", - "└─────────┴─────┘" - ] - }, - "execution_count": 67, - "metadata": {}, - "output_type": "execute_result" - } + }, + "execution_count": 34, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "lazy_plan = (\n", + " pl.scan_csv(data_dir / 'NCSU_Mascots_v1.csv')\n", + " .filter(pl.col('Species').is_not_null())\n", + " .group_by('Species')\n", + " .agg(pl.len().alias('n'))\n", + " .sort('n', descending=True)\n", + ")\n", + "\n", + "lazy_plan # shows a LazyFrame (a plan)" + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "id": "4d489f20", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'SORT BY [descending: [true]] [col(\"n\")]\\n AGGREGATE[maintain_order: false]\\n [len().alias(\"n\")] BY [col(\"Species\")]\\n FROM\\n Csv SCAN [data/NCSU_Mascots_v1.csv]\\n PROJECT 1/22 COLUMNS\\n SELECTION: col(\"Species\").is_not_null()\\n ESTIMATED ROWS: 59'" + ] + }, + "execution_count": 35, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "lazy_plan.explain()" + ] + }, + { + "cell_type": "code", + "execution_count": 36, + "id": "a18abc8f", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "shape: (10, 2)
Speciesn
stru32
"Dog"10
"Cat"5
"Bird"4
"Wolf"3
"Lizard"3
"Horse"3
"Rabbit"3
"Turtle"2
"Hamster"2
"Fish"2
" ], - "source": [ - "lazy_plan.collect().head(10)" - ] - }, - { - "cell_type": "markdown", - "id": "76dc5798", - "metadata": {}, - "source": [ - "## 11. (OPTIONAL) Polars native plotting\n", - "\n", - "Polars also supports native plotting via `DataFrame.plot` / `Series.plot` (Altair backend).\n", - "\n", - "This is useful when you want a quick chart directly from a Polars table without writing matplotlib code.\n", - "\n", - "If `altair` is not installed in your environment, skip this section.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 68, - "id": "2d2b1235", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Optional section skipped: install altair to use Polars native plotting.\n" - ] - } + "text/plain": [ + "shape: (10, 2)\n", + "┌─────────┬─────┐\n", + "│ Species ┆ n │\n", + "│ --- ┆ --- │\n", + "│ str ┆ u32 │\n", + "╞═════════╪═════╡\n", + "│ Dog ┆ 10 │\n", + "│ Cat ┆ 5 │\n", + "│ Bird ┆ 4 │\n", + "│ Wolf ┆ 3 │\n", + "│ Lizard ┆ 3 │\n", + "│ Horse ┆ 3 │\n", + "│ Rabbit ┆ 3 │\n", + "│ Turtle ┆ 2 │\n", + "│ Hamster ┆ 2 │\n", + "│ Fish ┆ 2 │\n", + "└─────────┴─────┘" + ] + }, + "execution_count": 36, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "lazy_plan.collect().head(10)" + ] + }, + { + "cell_type": "markdown", + "id": "76dc5798", + "metadata": {}, + "source": [ + "## 11. (OPTIONAL) Polars native plotting\n", + "\n", + "Polars also supports native plotting via `DataFrame.plot` / `Series.plot` (Altair backend).\n", + "\n", + "This is useful when you want a quick chart directly from a Polars table without writing matplotlib code.\n", + "\n", + "If `altair` is not installed in your environment, skip this section.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 43, + "id": "2d2b1235", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "\n", + "\n", + "
\n", + "" ], - "source": [ - "try:\n", - " import altair as alt\n", - "except ImportError:\n", - " print(\"Optional section skipped: install altair to use Polars native plotting.\")\n", - "else:\n", - " species_counts = (\n", - " mascots\n", - " .group_by('Species')\n", - " .len()\n", - " .sort('len', descending=True)\n", - " .head(10)\n", - " )\n", - "\n", - " species_counts.plot.bar(\n", - " x='Species',\n", - " y='len',\n", - " title='Top mascot species (Polars native plot)'\n", - " )\n" + "text/plain": [ + "alt.Chart(...)" ] - }, - { - "cell_type": "markdown", - "id": "2a30becb", - "metadata": {}, - "source": [ - "## Wrap-up\n", - "\n", - "What you practiced:\n", - "\n", - "- Profiling: shape, schema, preview, missingness (counts + rates), unique counts, duplicates\n", - "- Univariate + bivariate exploration with quick plots\n", - "- Cleaning + feature creation **before** aggregation\n", - "- `group_by().agg()` summaries (lossy)\n", - "- Window functions (group context without collapsing)\n", - "- Optional: pivoting, joins, and lazy execution\n", - "\n", - "Before you leave: write down **3 findings** and **1 next step**.\n" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.12.12" + }, + "metadata": {}, + "output_type": "display_data" } + ], + "source": [ + "try:\n", + " import altair as alt\n", + "except ImportError:\n", + " print(\"Optional section skipped: install altair to use Polars native plotting.\")\n", + "else:\n", + " species_counts = (\n", + " mascots\n", + " .group_by('Species')\n", + " .len()\n", + " .sort('len', descending=True)\n", + " .head(10)\n", + " )\n", + "\n", + " barplot_mascot = species_counts.plot.bar(\n", + " x='Species',\n", + " y='len'\n", + " ).properties(title='Top mascot species (Polars native plot)')\n", + " \n", + " barplot_mascot.show()\n" + ] + }, + { + "cell_type": "markdown", + "id": "2a30becb", + "metadata": {}, + "source": [ + "## Wrap-up\n", + "\n", + "What you practiced:\n", + "\n", + "- Profiling: shape, schema, preview, missingness (counts + rates), unique counts, duplicates\n", + "- Univariate + bivariate exploration with quick plots\n", + "- Cleaning + feature creation **before** aggregation\n", + "- `group_by().agg()` summaries (lossy)\n", + "- Window functions (group context without collapsing)\n", + "- Optional: pivoting, joins, and lazy execution\n", + "\n", + "Before you leave: write down **3 findings** and **1 next step**.\n" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" }, - "nbformat": 4, - "nbformat_minor": 5 + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.13.9" + } + }, + "nbformat": 4, + "nbformat_minor": 5 }