diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md deleted file mode 100644 index dd12a2a4..00000000 --- a/CONTRIBUTING.md +++ /dev/null @@ -1,37 +0,0 @@ -# Contributing to Learn.co Curriculum - -We're really exited that you're about to contribute to the [open curriculum](https://learn.co/content-license) on [Learn.co](https://learn.co). If this is your first time contributing, please continue reading to learn how to make the most meaningful and useful impact possible. - -## Raising an Issue to Encourage a Contribution - -If you notice a problem with the curriculum that you believe needs improvement -but you're unable to make the change yourself, you should raise a Github issue -containing a clear description of the problem. Include relevant snippets of -the content and/or screenshots if applicable. Curriculum owners regularly review -issue lists and your issue will be prioritized and addressed as appropriate. - -## Submitting a Pull Request to Suggest an Improvement - -If you see an opportunity for improvement and can make the change yourself go -ahead and use a typical git workflow to make it happen: - -* Fork this curriculum repository -* Make the change on your fork, with descriptive commits in the standard format -* Open a Pull Request against this repo - -A curriculum owner will review your change and approve or comment on it in due -course. - -# Why Contribute? - -Curriculum on Learn is publicly and freely available under Learn's -[Educational Content License](https://learn.co/content-license). By -embracing an open-source contribution model, our goal is for the curriculum -on Learn to become, in time, the best educational content the world has -ever seen. - -We need help from the community of Learners to maintain and improve the -educational content. Everything from fixing typos, to correcting -out-dated information, to improving exposition, to adding better examples, -to fixing tests—all contributions to making the curriculum more effective are -welcome. diff --git a/LICENSE.md b/LICENSE.md deleted file mode 100644 index 8c4ad9a7..00000000 --- a/LICENSE.md +++ /dev/null @@ -1,7 +0,0 @@ -#Learn.co Educational Content License - -Copyright (c) 2015 Flatiron School, Inc - -The Flatiron School, Inc. owns this Educational Content. However, the Flatiron School supports the development and availability of educational materials in the public domain. Therefore, the Flatiron School grants Users of the Flatiron Educational Content set forth in this repository certain rights to reuse, build upon and share such Educational Content subject to the terms of the Educational Content License set forth [here](http://learn.co/content-license) (http://learn.co/content-license). You must read carefully the terms and conditions contained in the Educational Content License as such terms govern access to and use of the Educational Content. - -Flatiron School is willing to allow you access to and use of the Educational Content only on the condition that you accept all of the terms and conditions contained in the Educational Content License set forth [here](http://learn.co/content-license) (http://learn.co/content-license). By accessing and/or using the Educational Content, you are agreeing to all of the terms and conditions contained in the Educational Content License. If you do not agree to any or all of the terms of the Educational Content License, you are prohibited from accessing, reviewing or using in any way the Educational Content. \ No newline at end of file diff --git a/README.md b/README.md index 5dd0f84d..2e16e68d 100644 --- a/README.md +++ b/README.md @@ -1,285 +1,36 @@ -# Phase 2 Project Description +# KING COUNTY REALTY ANALYSIS -Another module down - you're almost half way there! +## Overview -![awesome](https://raw.githubusercontent.com/learn-co-curriculum/dsc-phase-2-project-v2-3/main/halfway-there.gif) +The real estate sector encompasses a diverse range of activities involving the acquisition, development, management, and transaction of properties, including residential, commercial, and investment real estate. The urban real estate investment business model is undergoing a fundamental overhaul attributed to digitization and a growing market for smart and environmentally demanding buildings. -All that remains in Phase 2 is to put your newfound data science skills to use with a large project! -In this project description, we will cover: +## Business Understanding -* Project Overview: the project goal, audience, and dataset -* Deliverables: the specific items you are required to produce for this project -* Grading: how your project will be scored -* Getting Started: guidance for how to begin working +The client is King County Realty. A real estate company looking to understand the anatomy of a high priced home. Armed with a dataset of house sales in King County, the team of analysts look to define the most sought after characteristics in a home. -## Project Overview +This analysis will focus on the following factors that majorly affect the real estate industry to achieve the client's request:
+i) Market Trends: This analysis will help the client make well-informed decisions about the purchase, sale, and investment strategies of real estate. The dataset provides insights into pricing dynamics, market trends, and property features.
+ii) Customer Preferences: Understanding of consumer preferences and market demand through analysis of bedroom and bathroom numbers, property condition, view ratings makes targeted markeing and property customization easier and more scalable so as to suit all their clients' expectations.
+iii) Investment Opportunities: By analyzing variables including property size, location, condition, and market trends, investors in real estate can find properties that are cheap, evaluate possible returns, and optimize their investment portfolios.
+iv) Competitive Analysis: Real estate agents can benchmark against rivals, spot market gaps, and more by comparing property prices, sizes, amenities, and location features and set themselves out from the competition to obtain a market advantage. -For this project, you will use multiple linear regression modeling to analyze house sales in a northwestern county. +## Data Understatnding -### Business Problem +This project uses the King County House Sales dataset, which can be found in `kc_house_data.csv` in the data folder in this assignment's GitHub repository. The description of the column names can be found in `column_names.md` in the same folder. -It is up to you to define a stakeholder and business problem appropriate to this dataset. +### Challenges +Missing Values: +waterfront, view, and yr_renovated columns have missing values. This can be seen in the count row of the descriptive statistics section. -If you are struggling to define a stakeholder, we recommend you complete a project for a real estate agency that helps homeowners buy and/or sell homes. A business problem you could focus on for this stakeholder is the need to provide advice to homeowners about how home renovations might increase the estimated value of their homes, and by what amount. +Potential Data Quality Issues: +The bedrooms column has a maximum value of 33, which seems unusually high and might be an error or outlier. +The bathrooms column has a maximum value of 8, which could also be considered high and should be examined for outliers. +The yr_renovated column has a maximum value of 2015, which seems unusual as it's the same as the maximum value of the yr_built column. This suggests that some values in the yr_renovated column might represent the year built instead of renovation years. -### The Data +Inconsistent Data Types: +Some columns, such as waterfront, view, condition, and grade, appear to have categorical data but are represented as objects (strings) instead of categorical data types. +The date column is represented as an object (string) but should be converted to a datetime data type for easier manipulation and analysis. -This project uses the King County House Sales dataset, which can be found in `kc_house_data.csv` in the data folder in this assignment's GitHub repository. The description of the column names can be found in `column_names.md` in the same folder. As with most real world data sets, the column names are not perfectly described, so you'll have to do some research or use your best judgment if you have questions about what the data means. - -It is up to you to decide what data from this dataset to use and how to use it. If you are feeling overwhelmed or behind, we recommend you **ignore** some or all of the following features: - -* `date` -* `view` -* `sqft_above` -* `sqft_basement` -* `yr_renovated` -* `zipcode` -* `lat` -* `long` -* `sqft_living15` -* `sqft_lot15` - -### Key Points - -* **Your goal in regression modeling is to yield findings to support relevant recommendations. Those findings should include a metric describing overall model performance as well as at least two regression model coefficients.** As you explore the data and refine your stakeholder and business problem definitions, make sure you are also thinking about how a linear regression model adds value to your analysis. "The assignment was to use linear regression" is not an acceptable answer! You can also use additional statistical techniques other than linear regression, so long as you clearly explain why you are using each technique. - -* **You should demonstrate an iterative approach to modeling.** This means that you must build multiple models. Begin with a basic model, evaluate it, and then provide justification for and proceed to a new model. After you finish refining your models, you should provide 1-3 paragraphs in the notebook discussing your final model. - -* **Data visualization and analysis are no longer explicit project requirements, but they are still very important.** In Phase 1, your project stopped earlier in the CRISP-DM process. Now you are going a step further, to modeling. Data visualization and analysis will help you build better models and tell a better story to your stakeholders. - -## Deliverables - -There are three deliverables for this project: - -* A **non-technical presentation** -* A **Jupyter Notebook** -* A **GitHub repository** - -The deliverables requirements are almost the same as in the Phase 1 Project, and you can review those extended descriptions [here](https://github.com/learn-co-curriculum/dsc-phase-1-project-v2-3#deliverables). In general, everything is the same except the "Data Visualization" and "Data Analysis" requirements have been replaced by "Modeling" and "Regression Results" requirements. - -### Non-Technical Presentation - -Recall that the non-technical presentation is a slide deck presenting your analysis to ***business stakeholders***, and should be presented live as well as submitted in PDF form on Canvas. - -We recommend that you follow this structure, although the slide titles should be specific to your project: - -1. Beginning - - Overview - - Business and Data Understanding -2. Middle - - **Modeling** - - **Regression Results** -3. End - - Recommendations - - Next Steps - - Thank you - -Make sure that your discussion of modeling and regression results is geared towards a non-technical audience! Assume that their prior knowledge of regression modeling is minimal. You don't need to explain how linear regression works, but you should explain why linear regression is useful for the problem context. Make sure you translate any metrics or coefficients into their plain language implications. - -The graded elements for the non-technical presentation are the same as in [Phase 1](https://github.com/learn-co-curriculum/dsc-phase-1-project-v2-3#deliverables). - -### Jupyter Notebook - -Recall that the Jupyter Notebook is a notebook that uses Python and Markdown to present your analysis to a ***data science audience***. You will submit the notebook in PDF format on Canvas as well as in `.ipynb` format in your GitHub repository. - -The graded elements for the Jupyter Notebook are: - -* Business Understanding -* Data Understanding -* Data Preparation -* **Modeling** -* **Regression Results** -* Code Quality - -### GitHub Repository - -Recall that the GitHub repository is the cloud-hosted directory containing all of your project files as well as their version history. - -The requirements are the same as in [Phase 1](https://github.com/learn-co-curriculum/dsc-phase-1-project-v2-3#github-repository), except for the required sections in the `README.md`. - -For this project, the `README.md` file should contain: - -* Overview -* Business and Data Understanding - * Explain your stakeholder audience here -* **Modeling** -* **Regression Results** -* Conclusion - -Just like in Phase 1, the `README.md` file should be the bridge between your non technical presentation and the Jupyter Notebook. It should not contain the code used to develop your analysis, but should provide a more in-depth explanation of your methodology and analysis than what is described in your presentation slides. - -## Grading - -***To pass this project, you must pass each project rubric objective.*** The project rubric objectives for Phase 2 are: - -1. Attention to Detail -2. Statistical Communication -3. Data Preparation Fundamentals -4. Linear Modeling - -### Attention to Detail - -Just like in Phase 1, this rubric objective is based on your completion of checklist items. ***In Phase 2, you need to complete 70% (7 out of 10) or more of the checklist elements in order to pass the Attention to Detail objective.*** - -**NOTE THAT THE PASSING BAR IS HIGHER IN PHASE 2 THAN IT WAS IN PHASE 1!** - -The standard will increase with each Phase, until you will be required to complete all elements to pass Phase 5 (Capstone). - -#### Exceeds Objective - -80% or more of the project checklist items are complete - -#### Meets Objective (Passing Bar) - -70% of the project checklist items are complete - -#### Approaching Objective - -60% of the project checklist items are complete - -#### Does Not Meet Objective - -50% or fewer of the project checklist items are complete - -### Statistical Communication - -Recall that communication is one of the key data science "soft skills". In Phase 2, we are specifically focused on Statistical Communication. We define Statistical Communication as: - -> Communicating **results of statistical analyses** to diverse audiences via writing and live presentation - -Note that this is the same as in Phase 1, except we are replacing "basic data analysis" with "statistical analyses". - -High-quality Statistical Communication includes rationale, results, limitations, and recommendations: - -* **Rationale:** Explaining why you are using statistical analyses rather than basic data analysis - * For example, why are you using regression coefficients rather than just a graph? - * What about the problem or data is suitable for this form of analysis? - * For a data science audience, this includes your reasoning for the changes you applied while iterating between models. -* **Results:** Describing the overall model metrics and feature coefficients - * You need at least one overall model metric (e.g. r-squared or RMSE) and at least two feature coefficients. - * For a business audience, make sure you connect any metrics to real-world implications. You do not need to get into the details of how linear regression works. - * For a data science audience, you don't need to explain what a metric is, but make sure you explain why you chose that particular one. -* **Limitations:** Identifying the limitations and/or uncertainty present in your analysis - * This could include p-values/alpha values, confidence intervals, assumptions of linear regression, missing data, etc. - * In general, this should be more in-depth for a data science audience and more surface-level for a business audience. -* **Recommendations:** Interpreting the model results and limitations in the context of the business problem - * What should stakeholders _do_ with this information? - -#### Exceeds Objective - -Communicates the rationale, results, limitations, and specific recommendations of statistical analyses - -> See above for extended explanations of these terms. - -#### Meets Objective (Passing Bar) - -Successfully communicates the results of statistical analyses without any major errors - -> The minimum requirement is to communicate the _results_, meaning at least one overall model metric (e.g. r-squared or RMSE) as well as at least two feature coefficients. See the Approaching Objective section for an explanation of what a "major error" means. - -#### Approaching Objective - -Communicates the results of statistical analyses with at least one major error - -> A major error means that some aspect of your explanation is fundamentally incorrect. For example, if a feature coefficient is negative and you say that an increase in that feature results in an increase of the target, that would be a major error. Another example would be if you say that the feature with the highest coefficient is the "most statistically significant" while ignoring the p-value. One more example would be reporting a coefficient that is not statistically significant, rather than saying "no statistically significant linear relationship was found" - -> "**If a coefficient's t-statistic is not significant, don't interpret it at all.** You can't be sure that the value of the corresponding parameter in the underlying regression model isn't really zero." _DeVeaux, Velleman, and Bock (2012), Stats: Data and Models, 3rd edition, pg. 801_. Check out [this website](https://web.ma.utexas.edu/users/mks/statmistakes/TOC.html) for extensive additional examples of mistakes using statistics. - -> The easiest way to avoid making a major error is to have someone double-check your work. Reach out to peers on Slack and ask them to confirm whether your interpretation makes sense! - -#### Does Not Meet Objective - -Does not communicate the results of statistical analyses - -> It is not sufficient to just display the entire results summary. You need to pull out at least one overall model metric (e.g. r-squared, RMSE) and at least two feature coefficients, and explain what those numbers mean. - -### Data Preparation Fundamentals - -We define this objective as: - -> Applying appropriate **preprocessing** and feature engineering steps to tabular data in preparation for statistical modeling - -The two most important components of preprocessing for the Phase 2 project are: - -* **Handling Missing Values:** Missing values may be present in the features you want to use, either encoded as `NaN` or as some other value such as `"?"`. Before you can build a linear regression model, make sure you identify and address any missing values using techniques such as dropping or replacing data. -* **Handling Non-Numeric Data:** A linear regression model needs all of the features to be numeric, not categorical. For this project, ***be sure to pick at least one non-numeric feature and try including it in a model.*** You can identify that a feature is currently non-numeric if the type is `object` when you run `.info()` on your dataframe. Once you have identified the non-numeric features, address them using techniques such as ordinal or one-hot (dummy) encoding. - -There is no single correct way to handle either of these situations! Use your best judgement to decide what to do, and be sure to explain your rationale in the Markdown of your notebook. - -Feature engineering is encouraged but not required for this project. - -#### Exceeds Objective - -Goes above and beyond with data preparation, such as feature engineering or merging in outside datasets - -> One example of feature engineering could be using the `date` feature to create a new feature called `season`, which represents whether the home was sold in Spring, Summer, Fall, or Winter. - -> One example of merging in outside datasets could be finding data based on ZIP Code, such as household income or walkability, and joining that data with the provided CSV. - -#### Meets Objective (Passing Bar) - -Successfully prepares data for modeling, including converting at least one non-numeric feature into ordinal or binary data and handling missing data as needed - -> As a reminder, you can identify the non-numeric features by calling `.info()` on the dataframe and looking for type `object`. - -> Your final model does not necessarily need to include any features that were originally non-numeric, but you need to demonstrate your ability to handle this type of data. - -#### Approaching Objective - -Prepares some data successfully, but is unable to utilize non-numeric data - -> If you simply subset the dataframe to only columns with type `int64` or `float64`, your model will run, but you will not pass this objective. - -#### Does Not Meet Objective - -Does not prepare data for modeling - -### Linear Modeling - -According to [Kaggle's 2020 State of Data Science and Machine Learning Survey](https://www.kaggle.com/kaggle-survey-2020), linear and logistic regression are the most popular machine learning algorithms, used by 83.7% of data scientists. They are small, fast models compared to some of the models you will learn later, but have limitations in the kinds of relationships they are able to learn. - -In this project you are required to use linear regression as the primary statistical analysis, although you are free to use additional statistical techniques as appropriate. - -#### Exceeds Objective - -Goes above and beyond in the modeling process, such as recursive feature selection - -#### Meets Objective (Passing Bar) - -Successfully builds a baseline model as well as at least one iterated model, and correctly extracts insights from a final model without any major errors - -> We are looking for you to (1) create a baseline model, (2) iterate on that model, making adjustments that are supported by regression theory or by descriptive analysis of the data, and (3) select a final model and report on its metrics and coefficients - -> Ideally you would include written justifications for each model iteration, but at minimum the iterations must be _justifiable_ - -> For an explanation of "major errors", see the description below - -#### Approaching Objective - -Builds multiple models with at least one major error - -> The number one major error to avoid is including the target as one of your features. For example, if the target is `price` you should NOT make a "price per square foot" feature, because that feature would not be available if you didn't already know the price. - -> Other examples of major errors include: using a target other than `price`, attempting only simple linear regression (not multiple linear regression), dropping multiple one-hot encoded columns without explaining the resulting baseline, or using a unique identifier (`id` in this dataset) as a feature. - -#### Does Not Meet Objective - -Does not build multiple linear regression models - -## Getting Started - -Please start by reviewing the contents of this project description. If you have any questions, please ask your instructor ASAP. - -Next, you will need to complete the [***Project Proposal***](#project_proposal) which must be reviewed by your instructor before you can continue with the project. - -Here are some suggestions for creating your GitHub repository: - -1. Fork the [Phase 2 Project Repository](https://github.com/learn-co-curriculum/dsc-phase-2-project-v2-3), clone it locally, and work in the `student.ipynb` file. Make sure to also add and commit a PDF of your presentation to your repository with a file name of `presentation.pdf`. -2. Or, create a new repository from scratch by going to [github.com/new](https://github.com/new) and copying the data files from the Phase 2 Project Repository into your new repository. - - Recall that you can refer to the [Phase 1 Project Template](https://github.com/learn-co-curriculum/dsc-project-template) as an example structure - - This option will result in the most professional-looking portfolio repository, but can be more complicated to use. So if you are getting stuck with this option, try forking the project repository instead - -## Summary - -This is your first modeling project! Take what you have learned in Phase 2 to create a project with a more sophisticated analysis than you completed in Phase 1. You will build on these skills as we move into the predictive machine learning mindset in Phase 3. You've got this! +## Findings +The biggest factor to increase a house's price was number of floors. Adding $19,000 to the asking price for every additional floor. Other major contributing factors were number o bathrooms and the condition rating of the home. Houses with a view also demand a higher fee. \ No newline at end of file diff --git a/halfway-there.gif b/halfway-there.gif deleted file mode 100644 index 11d3c542..00000000 Binary files a/halfway-there.gif and /dev/null differ diff --git a/presentation.pdf b/presentation.pdf new file mode 100644 index 00000000..9a71cf96 Binary files /dev/null and b/presentation.pdf differ diff --git a/student.ipynb b/student.ipynb index d3bb34af..389d316c 100644 --- a/student.ipynb +++ b/student.ipynb @@ -4,23 +4,4004 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Final Project Submission\n", + "## King County Realty Analysis\n", "\n", - "Please fill out:\n", - "* Student name: \n", - "* Student pace: self paced / part time / full time\n", - "* Scheduled project review date/time: \n", - "* Instructor name: \n", - "* Blog post URL:\n" + "* Student names: Ian Musau, Mathew Karani, Jacinta Chepkemoi, Christine Malinga and Tabitha Berum.\n", + "* Student pace: Full time\n", + "* Instructor name: Nikita Njoroge\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Business Problem\n", + "\n", + "#### Overview\n", + "The client is King County Realty. A real estate company looking to understand the anatomy of a high priced home. Armed with a dataset of house sales in King County, the team of analysts look to define the most sought after characteristics in a home.\n", + "\n", + "This analysis will focus on the following factors that majorly affect the real estate industry to achieve the client's request:
\n", + "- Market Trends: This analysis will help the client make well-informed decisions about the purchase, sale, and investment strategies of real estate. The dataset provides insights into pricing dynamics, market trends, and property features.
\n", + "- Customer Preferences: Understanding of consumer preferences and market demand through analysis of bedroom and bathroom numbers, property condition, view ratings makes targeted markeing and property customization easier and more scalable so as to suit all their clients' expectations.
\n", + "- Investment Opportunities: By analyzing variables including property size, location, condition, and market trends, investors in real estate can find properties that are cheap, evaluate possible returns, and optimize their investment portfolios.
\n", + "- Competitive Analysis: Real estate agents can benchmark against rivals, spot market gaps, and more by comparing property prices, sizes, amenities, and location features and set themselves out from the competition to obtain a market advantage." + ] + }, + { + "cell_type": "code", + "execution_count": 267, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Data Understanding" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The dataset \"data_sourcing\" contains a plethora of information about real estate properties, such as property IDs, price ranges, dates of sales, number of bedrooms and bathrooms, living space and lot sizes, floor counts, waterfront access, view ratings, property condition and grade, years of construction and renovation, geographic coordinates, and information about competitive neighborhoods.\n", + "\n", + "Containing 21 columns and over 21,000 items, the dataset offers a plethora of data for investigating correlations, trends, and patterns in the real estate market. The diversity of property variables and the quantity of unique values in each column reflect the heterogeneity and complexity of the real estate landscape represented in the dataset.\n", + "\n", + "Gaining an understanding of the differences included in each characteristic and how they are distributed is essential for doing important analyses, such as identifying the factors influencing property values, assessing neighborhood characteristics, and predicting market trends. Furthermore, it will be crucial to address data quality issues like missing values, outliers, and inconsistent data types in order to ensure the accuracy and reliability of any studies or models based on this dataset." + ] + }, + { + "cell_type": "code", + "execution_count": 268, + "metadata": {}, + "outputs": [], + "source": [ + "class DataSourcing:\n", + " def __init__(self,dataframe):\n", + " self.original = dataframe\n", + " self.dataframe = dataframe\n", + " \n", + " def give_info(self):\n", + " message = f\"\"\"\n", + " ----------------------------------------------------------------------->\n", + " DESCRIPTION OF THE DATAFRAME IN QUESTION:\n", + " ----------------------------------------------------------------------->\n", + " \n", + " Dataframe information => {self.dataframe.info()}\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " \n", + " Dataframe shape => {self.dataframe.shape[0]} rows, {self.dataframe.shape[1]} columns\n", + " -------------------------------------------------------------------------------------------------------------------------> \n", + " \n", + " There are {len(self.dataframe.columns)} columns, namely: {self.dataframe.columns}. \n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " \n", + " The first 5 records in the dataframe are seen here:\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " {self.dataframe.head()}\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " \n", + " The last 5 records in the self.dataframe are as follows: \n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " {self.dataframe.tail()}\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " \n", + " The descriptive statistics of the dataframe (mean,median, max, min, std) are as follows:\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " {self.dataframe.describe()}\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " \"\"\"\n", + " print (message)\n", + " \n", + " def null_count(self):\n", + " return self.dataframe.isnull().sum()\n", + " \n", + " def unique_count(self):\n", + " return self.dataframe.nunique()\n", + " \n", + " def unique_per_column(self):\n", + " print(\"<----- UNIQUE VALUES IN EACH COLUMN ----->\")\n", + " for col in self.dataframe.columns:\n", + " print(f\"{col} ({len(self.dataframe[col].unique())} unique)\\n {sorted(self.dataframe[col].unique())}\")\n", + " print()\n", + " print(\"<----- END OF UNIQUE VALUES IN EACH COLUMN ----->\")\n", + " return\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Load the CSV dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 269, + "metadata": {}, + "outputs": [], + "source": [ + "# Call the load_data() function to load the data\n", + "self = pd.read_csv('data/kc_house_data.csv')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now we can start the data understanding and pass the dataframe to our DataSourcing class" + ] + }, + { + "cell_type": "code", + "execution_count": 270, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 21597 entries, 0 to 21596\n", + "Data columns (total 21 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 21597 non-null int64 \n", + " 1 date 21597 non-null object \n", + " 2 price 21597 non-null float64\n", + " 3 bedrooms 21597 non-null int64 \n", + " 4 bathrooms 21597 non-null float64\n", + " 5 sqft_living 21597 non-null int64 \n", + " 6 sqft_lot 21597 non-null int64 \n", + " 7 floors 21597 non-null float64\n", + " 8 waterfront 19221 non-null object \n", + " 9 view 21534 non-null object \n", + " 10 condition 21597 non-null object \n", + " 11 grade 21597 non-null object \n", + " 12 sqft_above 21597 non-null int64 \n", + " 13 sqft_basement 21597 non-null object \n", + " 14 yr_built 21597 non-null int64 \n", + " 15 yr_renovated 17755 non-null float64\n", + " 16 zipcode 21597 non-null int64 \n", + " 17 lat 21597 non-null float64\n", + " 18 long 21597 non-null float64\n", + " 19 sqft_living15 21597 non-null int64 \n", + " 20 sqft_lot15 21597 non-null int64 \n", + "dtypes: float64(6), int64(9), object(6)\n", + "memory usage: 3.5+ MB\n", + "\n", + " ----------------------------------------------------------------------->\n", + " DESCRIPTION OF THE DATAFRAME IN QUESTION:\n", + " ----------------------------------------------------------------------->\n", + " \n", + " Dataframe information => None\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " \n", + " Dataframe shape => 21597 rows, 21 columns\n", + " -------------------------------------------------------------------------------------------------------------------------> \n", + " \n", + " There are 21 columns, namely: Index(['id', 'date', 'price', 'bedrooms', 'bathrooms', 'sqft_living',\n", + " 'sqft_lot', 'floors', 'waterfront', 'view', 'condition', 'grade',\n", + " 'sqft_above', 'sqft_basement', 'yr_built', 'yr_renovated', 'zipcode',\n", + " 'lat', 'long', 'sqft_living15', 'sqft_lot15'],\n", + " dtype='object'). \n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " \n", + " The first 5 records in the dataframe are seen here:\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "0 7129300520 10/13/2014 221900.0 3 1.00 1180 \n", + "1 6414100192 12/9/2014 538000.0 3 2.25 2570 \n", + "2 5631500400 2/25/2015 180000.0 2 1.00 770 \n", + "3 2487200875 12/9/2014 604000.0 4 3.00 1960 \n", + "4 1954400510 2/18/2015 510000.0 3 2.00 1680 \n", + "\n", + " sqft_lot floors waterfront view ... grade sqft_above \\\n", + "0 5650 1.0 NaN NONE ... 7 Average 1180 \n", + "1 7242 2.0 NO NONE ... 7 Average 2170 \n", + "2 10000 1.0 NO NONE ... 6 Low Average 770 \n", + "3 5000 1.0 NO NONE ... 7 Average 1050 \n", + "4 8080 1.0 NO NONE ... 8 Good 1680 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "0 0.0 1955 0.0 98178 47.5112 -122.257 \n", + "1 400.0 1951 1991.0 98125 47.7210 -122.319 \n", + "2 0.0 1933 NaN 98028 47.7379 -122.233 \n", + "3 910.0 1965 0.0 98136 47.5208 -122.393 \n", + "4 0.0 1987 0.0 98074 47.6168 -122.045 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "0 1340 5650 \n", + "1 1690 7639 \n", + "2 2720 8062 \n", + "3 1360 5000 \n", + "4 1800 7503 \n", + "\n", + "[5 rows x 21 columns]\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " \n", + " The last 5 records in the self.dataframe are as follows: \n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "21592 263000018 5/21/2014 360000.0 3 2.50 1530 \n", + "21593 6600060120 2/23/2015 400000.0 4 2.50 2310 \n", + "21594 1523300141 6/23/2014 402101.0 2 0.75 1020 \n", + "21595 291310100 1/16/2015 400000.0 3 2.50 1600 \n", + "21596 1523300157 10/15/2014 325000.0 2 0.75 1020 \n", + "\n", + " sqft_lot floors waterfront view ... grade sqft_above \\\n", + "21592 1131 3.0 NO NONE ... 8 Good 1530 \n", + "21593 5813 2.0 NO NONE ... 8 Good 2310 \n", + "21594 1350 2.0 NO NONE ... 7 Average 1020 \n", + "21595 2388 2.0 NaN NONE ... 8 Good 1600 \n", + "21596 1076 2.0 NO NONE ... 7 Average 1020 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "21592 0.0 2009 0.0 98103 47.6993 -122.346 \n", + "21593 0.0 2014 0.0 98146 47.5107 -122.362 \n", + "21594 0.0 2009 0.0 98144 47.5944 -122.299 \n", + "21595 0.0 2004 0.0 98027 47.5345 -122.069 \n", + "21596 0.0 2008 0.0 98144 47.5941 -122.299 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "21592 1530 1509 \n", + "21593 1830 7200 \n", + "21594 1020 2007 \n", + "21595 1410 1287 \n", + "21596 1020 1357 \n", + "\n", + "[5 rows x 21 columns]\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " \n", + " The descriptive statistics of the dataframe (mean,median, max, min, std) are as follows:\n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " id price bedrooms bathrooms sqft_living \\\n", + "count 2.159700e+04 2.159700e+04 21597.000000 21597.000000 21597.000000 \n", + "mean 4.580474e+09 5.402966e+05 3.373200 2.115826 2080.321850 \n", + "std 2.876736e+09 3.673681e+05 0.926299 0.768984 918.106125 \n", + "min 1.000102e+06 7.800000e+04 1.000000 0.500000 370.000000 \n", + "25% 2.123049e+09 3.220000e+05 3.000000 1.750000 1430.000000 \n", + "50% 3.904930e+09 4.500000e+05 3.000000 2.250000 1910.000000 \n", + "75% 7.308900e+09 6.450000e+05 4.000000 2.500000 2550.000000 \n", + "max 9.900000e+09 7.700000e+06 33.000000 8.000000 13540.000000 \n", + "\n", + " sqft_lot floors sqft_above yr_built yr_renovated \\\n", + "count 2.159700e+04 21597.000000 21597.000000 21597.000000 17755.000000 \n", + "mean 1.509941e+04 1.494096 1788.596842 1970.999676 83.636778 \n", + "std 4.141264e+04 0.539683 827.759761 29.375234 399.946414 \n", + "min 5.200000e+02 1.000000 370.000000 1900.000000 0.000000 \n", + "25% 5.040000e+03 1.000000 1190.000000 1951.000000 0.000000 \n", + "50% 7.618000e+03 1.500000 1560.000000 1975.000000 0.000000 \n", + "75% 1.068500e+04 2.000000 2210.000000 1997.000000 0.000000 \n", + "max 1.651359e+06 3.500000 9410.000000 2015.000000 2015.000000 \n", + "\n", + " zipcode lat long sqft_living15 sqft_lot15 \n", + "count 21597.000000 21597.000000 21597.000000 21597.000000 21597.000000 \n", + "mean 98077.951845 47.560093 -122.213982 1986.620318 12758.283512 \n", + "std 53.513072 0.138552 0.140724 685.230472 27274.441950 \n", + "min 98001.000000 47.155900 -122.519000 399.000000 651.000000 \n", + "25% 98033.000000 47.471100 -122.328000 1490.000000 5100.000000 \n", + "50% 98065.000000 47.571800 -122.231000 1840.000000 7620.000000 \n", + "75% 98118.000000 47.678000 -122.125000 2360.000000 10083.000000 \n", + "max 98199.000000 47.777600 -121.315000 6210.000000 871200.000000 \n", + " ------------------------------------------------------------------------------------------------------------------------->\n", + " \n" + ] + } + ], + "source": [ + "house_data = DataSourcing(dataframe=self)\n", + "house_data.give_info()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The insights derived from the give_info() method provide essential details about the dataset, contributing to a comprehensive understanding:\n", + "\n", + "This data set has 21 columns namely: id, date, price, bedrooms, bathrooms, sqft_living, sqft_lot, floors, waterfront, view, condition, grade, sqft_above, sqft_basement, yr_built, yr_renovated, zipcode, lat, long, sqft_living15, sqft_lot15.\n", + "\n", + "Missing Values: waterfront, view, and yr_renovated columns have missing values. This can be seen in the count row of the descriptive statistics section.\n", + "\n", + "Potential Data Quality Issues: The bedrooms column has a maximum value of 33, which seems unusually high and might be an error or outlier. The bathrooms column has a maximum value of 8, which could also be considered high and should be examined for outliers. The yr_renovated column has a maximum value of 2015, which seems unusual as it's the same as the maximum value of the yr_built column. This suggests that some values in the yr_renovated column might represent the year built instead of renovation years.\n", + "\n", + "Inconsistent Data Types: Some columns, such as waterfront, view, condition, and grade, appear to have categorical data but are represented as objects (strings) instead of categorical data types. The date column is represented as an object (string) but should be converted to a datetime data type for easier manipulation and analysis.\n", + "\n", + "Potential Outliers: Outliers may exist in numerical columns such as price, bedrooms, bathrooms, sqft_living, sqft_lot, sqft_above, sqft_basement, yr_built, yr_renovated, lat, long, sqft_living15, and sqft_lot15. Visualizing the distributions of these columns can help identify outliers.\n", + "\n", + "Zipcode as Numeric: The zipcode column is currently represented as an integer, but it might be more appropriate to treat it as a categorical variable since it represents different geographic areas.\n", + "\n", + "Inconsistent Naming Conventions: Some column names are in snake_case format (sqft_living, sqft_lot) while others are in camelCase (sqftLiving15, sqftLot15). It's best to use a consistent naming convention throughout the dataset." + ] + }, + { + "cell_type": "code", + "execution_count": 271, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "id 21420\n", + "date 372\n", + "price 3622\n", + "bedrooms 12\n", + "bathrooms 29\n", + "sqft_living 1034\n", + "sqft_lot 9776\n", + "floors 6\n", + "waterfront 2\n", + "view 5\n", + "condition 5\n", + "grade 11\n", + "sqft_above 942\n", + "sqft_basement 304\n", + "yr_built 116\n", + "yr_renovated 70\n", + "zipcode 70\n", + "lat 5033\n", + "long 751\n", + "sqft_living15 777\n", + "sqft_lot15 8682\n", + "dtype: int64" + ] + }, + "execution_count": 271, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Check for unique data so as to avoid redundancy in our dataset\n", + "house_data.unique_count()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Upon examining the data, it appears to be authentic and devoid of anomalies for several reasons:\n", + "\n", + "Variety in Property IDs: The dataset comprises 21,420 distinct IDs, each representing a unique property. Diverse Sale Dates: A total of 372 unique sale dates are recorded, reflecting the temporal spread of property transactions. Price Variation: With 3,622 unique price points, the dataset demonstrates a wide range of property prices. Bedroom and Bathroom Distribution: There are 12 distinct bedroom counts and 29 unique bathroom counts, indicating diverse property configurations. Living Space and Lot Sizes: The dataset encompasses 1,034 different living space sizes and 9,776 distinct lot sizes, showcasing the variability in property dimensions. Floor Counts: Properties are characterized by 6 unique floor counts, illustrating differences in architectural design. Waterfront Access: The \"waterfront\" column exhibits 2 unique values, indicating the presence or absence of waterfront access for properties. View Ratings: Properties are rated on 5 unique view levels, reflecting varying scenic qualities. Condition and Grade: The dataset includes 5 distinct property conditions and 11 unique grade categories, offering insights into property quality. Basement and Above-Ground Area: Property basements and above-ground areas are characterized by 304 and 942 unique size values, respectively. Construction and Renovation Years: There are 116 unique construction years and 70 distinct renovation years, providing historical context for properties. Geographical Distribution: Properties are located in 70 unique zip codes, with 5,033 distinct latitude and 751 longitude values, showcasing geographical diversity. Neighborhood Comparison: The dataset features 777 different living space sizes and 8,682 lot sizes for the 15 nearest neighbors, enabling neighborhood-specific analyses.\n", + "\n", + "The dataset exhibits substantial diversity across various property attributes, including size, location, condition, and price. These unique counts offer valuable insights into the dataset's composition, facilitating informed analyses and decision-making processes within the real estate domain." + ] + }, + { + "cell_type": "code", + "execution_count": 272, + "metadata": {}, + "outputs": [], + "source": [ + "def clean_data(data):\n", + " # Handling missing values\n", + " data.dropna(inplace=True) # Remove rows with any missing values\n", + "\n", + " # Removing duplicates\n", + " data.drop_duplicates(inplace=True)\n", + "\n", + " # Standardizing data formats (if needed)\n", + " # For example, converting categorical variables to a consistent format\n", + "\n", + " # Reset index after dropping rows\n", + " data.reset_index(drop=True, inplace=True)\n", + " \n", + " return data" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Data Preparation\n" + ] + }, + { + "cell_type": "code", + "execution_count": 273, + "metadata": {}, + "outputs": [], + "source": [ + "# import necessary libraries\n", + "import pandas as pd\n", + "import numpy as np\n" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
iddatepricebedroomsbathroomssqft_livingsqft_lotfloorswaterfrontview...gradesqft_abovesqft_basementyr_builtyr_renovatedzipcodelatlongsqft_living15sqft_lot15
0712930052010/13/2014221900.031.00118056501.0NaNNONE...7 Average11800.019550.09817847.5112-122.25713405650
1641410019212/9/2014538000.032.25257072422.0NONONE...7 Average2170400.019511991.09812547.7210-122.31916907639
256315004002/25/2015180000.021.00770100001.0NONONE...6 Low Average7700.01933NaN9802847.7379-122.23327208062
3248720087512/9/2014604000.043.00196050001.0NONONE...7 Average1050910.019650.09813647.5208-122.39313605000
419544005102/18/2015510000.032.00168080801.0NONONE...8 Good16800.019870.09807447.6168-122.04518007503
\n", + "

5 rows × 21 columns

\n", + "
" + ], + "text/plain": [ + " id date price bedrooms bathrooms sqft_living \\\n", + "0 7129300520 10/13/2014 221900.0 3 1.00 1180 \n", + "1 6414100192 12/9/2014 538000.0 3 2.25 2570 \n", + "2 5631500400 2/25/2015 180000.0 2 1.00 770 \n", + "3 2487200875 12/9/2014 604000.0 4 3.00 1960 \n", + "4 1954400510 2/18/2015 510000.0 3 2.00 1680 \n", + "\n", + " sqft_lot floors waterfront view ... grade sqft_above \\\n", + "0 5650 1.0 NaN NONE ... 7 Average 1180 \n", + "1 7242 2.0 NO NONE ... 7 Average 2170 \n", + "2 10000 1.0 NO NONE ... 6 Low Average 770 \n", + "3 5000 1.0 NO NONE ... 7 Average 1050 \n", + "4 8080 1.0 NO NONE ... 8 Good 1680 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "0 0.0 1955 0.0 98178 47.5112 -122.257 \n", + "1 400.0 1951 1991.0 98125 47.7210 -122.319 \n", + "2 0.0 1933 NaN 98028 47.7379 -122.233 \n", + "3 910.0 1965 0.0 98136 47.5208 -122.393 \n", + "4 0.0 1987 0.0 98074 47.6168 -122.045 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "0 1340 5650 \n", + "1 1690 7639 \n", + "2 2720 8062 \n", + "3 1360 5000 \n", + "4 1800 7503 \n", + "\n", + "[5 rows x 21 columns]" + ] + }, + "execution_count": 3, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#Loading the dataset\n", + "house_data=pd.read_csv(\"data/kc_house_data.csv\")\n", + "house_data.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 21597 entries, 0 to 21596\n", + "Data columns (total 21 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 21597 non-null int64 \n", + " 1 date 21597 non-null object \n", + " 2 price 21597 non-null float64\n", + " 3 bedrooms 21597 non-null int64 \n", + " 4 bathrooms 21597 non-null float64\n", + " 5 sqft_living 21597 non-null int64 \n", + " 6 sqft_lot 21597 non-null int64 \n", + " 7 floors 21597 non-null float64\n", + " 8 waterfront 19221 non-null object \n", + " 9 view 21534 non-null object \n", + " 10 condition 21597 non-null object \n", + " 11 grade 21597 non-null object \n", + " 12 sqft_above 21597 non-null int64 \n", + " 13 sqft_basement 21597 non-null object \n", + " 14 yr_built 21597 non-null int64 \n", + " 15 yr_renovated 17755 non-null float64\n", + " 16 zipcode 21597 non-null int64 \n", + " 17 lat 21597 non-null float64\n", + " 18 long 21597 non-null float64\n", + " 19 sqft_living15 21597 non-null int64 \n", + " 20 sqft_lot15 21597 non-null int64 \n", + "dtypes: float64(6), int64(9), object(6)\n", + "memory usage: 3.5+ MB\n" + ] + } + ], + "source": [ + "#loading the summary metadata\n", + "house_data.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 276, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "No duplicates found.\n" + ] + } + ], + "source": [ + "#checking for duplicates\n", + "def check_duplicates(dataset):\n", + " duplicates = dataset.duplicated()\n", + " if duplicates.any():\n", + " duplicated_rows = dataset[duplicates]\n", + " print(\"Duplicate rows:\")\n", + " print(duplicated_rows)\n", + " else:\n", + " print(\"No duplicates found.\")\n", + "check_duplicates(house_data)" + ] + }, + { + "cell_type": "code", + "execution_count": 277, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Missing values found:\n", + "id 0\n", + "date 0\n", + "price 0\n", + "bedrooms 0\n", + "bathrooms 0\n", + "sqft_living 0\n", + "sqft_lot 0\n", + "floors 0\n", + "waterfront 2376\n", + "view 63\n", + "condition 0\n", + "grade 0\n", + "sqft_above 0\n", + "sqft_basement 0\n", + "yr_built 0\n", + "yr_renovated 3842\n", + "zipcode 0\n", + "lat 0\n", + "long 0\n", + "sqft_living15 0\n", + "sqft_lot15 0\n", + "dtype: int64\n" + ] + } + ], + "source": [ + "#checking for missing values\n", + "def check_missing_values(dataset):\n", + " missing_values = dataset.isnull().sum()\n", + " if missing_values.any():\n", + " print(\"Missing values found:\")\n", + " print(missing_values)\n", + " else:\n", + " print(\"No missing values found.\")\n", + "\n", + "check_missing_values(house_data)" + ] + }, + { + "cell_type": "code", + "execution_count": 278, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idpricebedroomsbathroomssqft_livingsqft_lotfloorssqft_aboveyr_builtyr_renovatedzipcodelatlongsqft_living15sqft_lot15
count2.159700e+042.159700e+0421597.00000021597.00000021597.0000002.159700e+0421597.00000021597.00000021597.00000017755.00000021597.00000021597.00000021597.00000021597.00000021597.000000
mean4.580474e+095.402966e+053.3732002.1158262080.3218501.509941e+041.4940961788.5968421970.99967683.63677898077.95184547.560093-122.2139821986.62031812758.283512
std2.876736e+093.673681e+050.9262990.768984918.1061254.141264e+040.539683827.75976129.375234399.94641453.5130720.1385520.140724685.23047227274.441950
min1.000102e+067.800000e+041.0000000.500000370.0000005.200000e+021.000000370.0000001900.0000000.00000098001.00000047.155900-122.519000399.000000651.000000
25%2.123049e+093.220000e+053.0000001.7500001430.0000005.040000e+031.0000001190.0000001951.0000000.00000098033.00000047.471100-122.3280001490.0000005100.000000
50%3.904930e+094.500000e+053.0000002.2500001910.0000007.618000e+031.5000001560.0000001975.0000000.00000098065.00000047.571800-122.2310001840.0000007620.000000
75%7.308900e+096.450000e+054.0000002.5000002550.0000001.068500e+042.0000002210.0000001997.0000000.00000098118.00000047.678000-122.1250002360.00000010083.000000
max9.900000e+097.700000e+0633.0000008.00000013540.0000001.651359e+063.5000009410.0000002015.0000002015.00000098199.00000047.777600-121.3150006210.000000871200.000000
\n", + "
" + ], + "text/plain": [ + " id price bedrooms bathrooms sqft_living \\\n", + "count 2.159700e+04 2.159700e+04 21597.000000 21597.000000 21597.000000 \n", + "mean 4.580474e+09 5.402966e+05 3.373200 2.115826 2080.321850 \n", + "std 2.876736e+09 3.673681e+05 0.926299 0.768984 918.106125 \n", + "min 1.000102e+06 7.800000e+04 1.000000 0.500000 370.000000 \n", + "25% 2.123049e+09 3.220000e+05 3.000000 1.750000 1430.000000 \n", + "50% 3.904930e+09 4.500000e+05 3.000000 2.250000 1910.000000 \n", + "75% 7.308900e+09 6.450000e+05 4.000000 2.500000 2550.000000 \n", + "max 9.900000e+09 7.700000e+06 33.000000 8.000000 13540.000000 \n", + "\n", + " sqft_lot floors sqft_above yr_built yr_renovated \\\n", + "count 2.159700e+04 21597.000000 21597.000000 21597.000000 17755.000000 \n", + "mean 1.509941e+04 1.494096 1788.596842 1970.999676 83.636778 \n", + "std 4.141264e+04 0.539683 827.759761 29.375234 399.946414 \n", + "min 5.200000e+02 1.000000 370.000000 1900.000000 0.000000 \n", + "25% 5.040000e+03 1.000000 1190.000000 1951.000000 0.000000 \n", + "50% 7.618000e+03 1.500000 1560.000000 1975.000000 0.000000 \n", + "75% 1.068500e+04 2.000000 2210.000000 1997.000000 0.000000 \n", + "max 1.651359e+06 3.500000 9410.000000 2015.000000 2015.000000 \n", + "\n", + " zipcode lat long sqft_living15 sqft_lot15 \n", + "count 21597.000000 21597.000000 21597.000000 21597.000000 21597.000000 \n", + "mean 98077.951845 47.560093 -122.213982 1986.620318 12758.283512 \n", + "std 53.513072 0.138552 0.140724 685.230472 27274.441950 \n", + "min 98001.000000 47.155900 -122.519000 399.000000 651.000000 \n", + "25% 98033.000000 47.471100 -122.328000 1490.000000 5100.000000 \n", + "50% 98065.000000 47.571800 -122.231000 1840.000000 7620.000000 \n", + "75% 98118.000000 47.678000 -122.125000 2360.000000 10083.000000 \n", + "max 98199.000000 47.777600 -121.315000 6210.000000 871200.000000 " + ] + }, + "execution_count": 278, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "#loading summary statistics for each column\n", + "house_data.describe()" + ] + }, + { + "cell_type": "code", + "execution_count": 279, + "metadata": {}, + "outputs": [], + "source": [ + "#dropping the rows with the missing values in the view column\n", + "def drop_missing_view_rows(dataset):\n", + " dataset.dropna(subset=['view'], inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 280, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 21597 entries, 0 to 21596\n", + "Data columns (total 21 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 21597 non-null int64 \n", + " 1 date 21597 non-null object \n", + " 2 price 21597 non-null float64\n", + " 3 bedrooms 21597 non-null int64 \n", + " 4 bathrooms 21597 non-null float64\n", + " 5 sqft_living 21597 non-null int64 \n", + " 6 sqft_lot 21597 non-null int64 \n", + " 7 floors 21597 non-null float64\n", + " 8 waterfront 19221 non-null object \n", + " 9 view 21534 non-null object \n", + " 10 condition 21597 non-null object \n", + " 11 grade 21597 non-null object \n", + " 12 sqft_above 21597 non-null int64 \n", + " 13 sqft_basement 21597 non-null object \n", + " 14 yr_built 21597 non-null int64 \n", + " 15 yr_renovated 17755 non-null float64\n", + " 16 zipcode 21597 non-null int64 \n", + " 17 lat 21597 non-null float64\n", + " 18 long 21597 non-null float64\n", + " 19 sqft_living15 21597 non-null int64 \n", + " 20 sqft_lot15 21597 non-null int64 \n", + "dtypes: float64(6), int64(9), object(6)\n", + "memory usage: 3.5+ MB\n" + ] + } + ], + "source": [ + "house_data.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 281, + "metadata": {}, + "outputs": [], + "source": [ + "#converting the waterfront and year renovated columns to integers\n", + "def convert_string_to_int(dataset, columns):\n", + " for column in columns:\n", + " dataset[column] = pd.to_numeric(dataset[column], errors='coerce')\n", + "columns_to_convert = ['waterfront', 'yr_renovated'] \n", + "convert_string_to_int(house_data, columns_to_convert)" + ] + }, + { + "cell_type": "code", + "execution_count": 282, + "metadata": {}, + "outputs": [], + "source": [ + "# Replace missing values in 'waterfront' and 'yr_renovated' columns with the mean\n", + "def replace_missing_with_mean_for_columns(dataset, columns):\n", + " for column in columns:\n", + " dataset[column].fillna(dataset[column].mean(), inplace=True)\n", + "\n", + "columns_to_replace = ['waterfront', 'yr_renovated'] \n", + "replace_missing_with_mean_for_columns(house_data, columns_to_replace)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 283, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 21597 entries, 0 to 21596\n", + "Data columns (total 21 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 21597 non-null int64 \n", + " 1 date 21597 non-null object \n", + " 2 price 21597 non-null float64\n", + " 3 bedrooms 21597 non-null int64 \n", + " 4 bathrooms 21597 non-null float64\n", + " 5 sqft_living 21597 non-null int64 \n", + " 6 sqft_lot 21597 non-null int64 \n", + " 7 floors 21597 non-null float64\n", + " 8 waterfront 0 non-null float64\n", + " 9 view 21534 non-null object \n", + " 10 condition 21597 non-null object \n", + " 11 grade 21597 non-null object \n", + " 12 sqft_above 21597 non-null int64 \n", + " 13 sqft_basement 21597 non-null object \n", + " 14 yr_built 21597 non-null int64 \n", + " 15 yr_renovated 21597 non-null float64\n", + " 16 zipcode 21597 non-null int64 \n", + " 17 lat 21597 non-null float64\n", + " 18 long 21597 non-null float64\n", + " 19 sqft_living15 21597 non-null int64 \n", + " 20 sqft_lot15 21597 non-null int64 \n", + "dtypes: float64(7), int64(9), object(5)\n", + "memory usage: 3.5+ MB\n" + ] + } + ], + "source": [ + "house_data.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 284, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 21597 entries, 0 to 21596\n", + "Data columns (total 21 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 21597 non-null int64 \n", + " 1 date 21597 non-null object \n", + " 2 price 21597 non-null float64\n", + " 3 bedrooms 21597 non-null int64 \n", + " 4 bathrooms 21597 non-null float64\n", + " 5 sqft_living 21597 non-null int64 \n", + " 6 sqft_lot 21597 non-null int64 \n", + " 7 floors 21597 non-null float64\n", + " 8 waterfront 0 non-null float64\n", + " 9 view 21534 non-null object \n", + " 10 condition 21597 non-null object \n", + " 11 grade 21597 non-null object \n", + " 12 sqft_above 21597 non-null int64 \n", + " 13 sqft_basement 21597 non-null object \n", + " 14 yr_built 21597 non-null int64 \n", + " 15 yr_renovated 21597 non-null float64\n", + " 16 zipcode 21597 non-null int64 \n", + " 17 lat 21597 non-null float64\n", + " 18 long 21597 non-null float64\n", + " 19 sqft_living15 21597 non-null int64 \n", + " 20 sqft_lot15 21597 non-null int64 \n", + "dtypes: float64(7), int64(9), object(5)\n", + "memory usage: 3.5+ MB\n" + ] + } + ], + "source": [ + "#reloading the data set\n", + "house_data.info()" + ] + }, + { + "cell_type": "code", + "execution_count": 285, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Index: 21534 entries, 0 to 21596\n", + "Data columns (total 21 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 21534 non-null int64 \n", + " 1 date 21534 non-null object \n", + " 2 price 21534 non-null float64\n", + " 3 bedrooms 21534 non-null int64 \n", + " 4 bathrooms 21534 non-null float64\n", + " 5 sqft_living 21534 non-null int64 \n", + " 6 sqft_lot 21534 non-null int64 \n", + " 7 floors 21534 non-null float64\n", + " 8 waterfront 0 non-null float64\n", + " 9 view 21534 non-null object \n", + " 10 condition 21534 non-null object \n", + " 11 grade 21534 non-null object \n", + " 12 sqft_above 21534 non-null int64 \n", + " 13 sqft_basement 21534 non-null object \n", + " 14 yr_built 21534 non-null int64 \n", + " 15 yr_renovated 21534 non-null float64\n", + " 16 zipcode 21534 non-null int64 \n", + " 17 lat 21534 non-null float64\n", + " 18 long 21534 non-null float64\n", + " 19 sqft_living15 21534 non-null int64 \n", + " 20 sqft_lot15 21534 non-null int64 \n", + "dtypes: float64(7), int64(9), object(5)\n", + "memory usage: 3.6+ MB\n", + "None\n" + ] + } + ], + "source": [ + "drop_missing_view_rows(house_data)\n", + "print(house_data.info())" + ] + }, + { + "cell_type": "code", + "execution_count": 286, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Index: 21534 entries, 0 to 21596\n", + "Data columns (total 20 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 21534 non-null int64 \n", + " 1 date 21534 non-null object \n", + " 2 price 21534 non-null float64\n", + " 3 bedrooms 21534 non-null int64 \n", + " 4 bathrooms 21534 non-null float64\n", + " 5 sqft_living 21534 non-null int64 \n", + " 6 sqft_lot 21534 non-null int64 \n", + " 7 floors 21534 non-null float64\n", + " 8 view 21534 non-null object \n", + " 9 condition 21534 non-null object \n", + " 10 grade 21534 non-null object \n", + " 11 sqft_above 21534 non-null int64 \n", + " 12 sqft_basement 21534 non-null object \n", + " 13 yr_built 21534 non-null int64 \n", + " 14 yr_renovated 21534 non-null float64\n", + " 15 zipcode 21534 non-null int64 \n", + " 16 lat 21534 non-null float64\n", + " 17 long 21534 non-null float64\n", + " 18 sqft_living15 21534 non-null int64 \n", + " 19 sqft_lot15 21534 non-null int64 \n", + "dtypes: float64(6), int64(9), object(5)\n", + "memory usage: 3.5+ MB\n", + "None\n" + ] + } + ], + "source": [ + "#Dropping the waterfront column since it contains missing values\n", + "house_data.drop('waterfront', axis=1, inplace=True)\n", + "\n", + "# Verify the changes\n", + "print(house_data.info())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Checking for outliers" + ] + }, + { + "cell_type": "code", + "execution_count": 287, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Empty DataFrame\n", + "Columns: [id, date, price, bedrooms, bathrooms, sqft_living, sqft_lot, floors, view, condition, grade, sqft_above, sqft_basement, yr_built, yr_renovated, zipcode, lat, long, sqft_living15, sqft_lot15]\n", + "Index: []\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "21 2524049179 8/26/2014 2000000.0 3 2.75 3050 \n", + "49 822039084 3/11/2015 1350000.0 3 2.50 2753 \n", + "69 1802000060 6/12/2014 1330000.0 5 2.25 3200 \n", + "125 4389200955 3/2/2015 1450000.0 4 2.75 2750 \n", + "153 7855801670 4/1/2015 2250000.0 4 3.25 5180 \n", + "... ... ... ... ... ... ... \n", + "21535 1561750040 12/24/2014 1380000.0 5 4.50 4350 \n", + "21552 524059330 1/30/2015 1700000.0 4 3.50 3830 \n", + "21560 9253900271 1/7/2015 3570000.0 5 4.50 4850 \n", + "21581 191100405 4/21/2015 1580000.0 4 3.25 3410 \n", + "21584 249000205 10/15/2014 1540000.0 5 3.75 4470 \n", + "\n", + " sqft_lot floors view condition grade sqft_above \\\n", + "21 44867 1.0 EXCELLENT Average 9 Better 2330 \n", + "49 65005 1.0 AVERAGE Very Good 9 Better 2165 \n", + "69 20158 1.0 NONE Average 8 Good 1600 \n", + "125 17789 1.5 NONE Average 8 Good 1980 \n", + "153 19850 2.0 GOOD Average 12 Luxury 3540 \n", + "... ... ... ... ... ... ... \n", + "21535 13405 2.0 NONE Average 11 Excellent 4350 \n", + "21552 8963 2.0 NONE Average 10 Very Good 3120 \n", + "21560 10584 2.0 EXCELLENT Average 10 Very Good 3540 \n", + "21581 10125 2.0 NONE Average 10 Very Good 3410 \n", + "21584 8088 2.0 NONE Average 11 Excellent 4470 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "21 720.0 1968 0.000000 98040 47.5316 -122.233 \n", + "49 588.0 1953 0.000000 98070 47.4041 -122.451 \n", + "69 1600.0 1965 0.000000 98004 47.6303 -122.215 \n", + "125 770.0 1914 1992.000000 98004 47.6141 -122.212 \n", + "153 1640.0 2006 0.000000 98006 47.5620 -122.162 \n", + "... ... ... ... ... ... ... \n", + "21535 0.0 2014 0.000000 98074 47.6018 -122.060 \n", + "21552 710.0 2014 0.000000 98004 47.5990 -122.197 \n", + "21560 1310.0 2007 0.000000 98008 47.5943 -122.110 \n", + "21581 ? 2007 83.636778 98040 47.5653 -122.223 \n", + "21584 0.0 2008 0.000000 98004 47.6321 -122.200 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "21 4110 20336 \n", + "49 2680 72513 \n", + "69 3390 20158 \n", + "125 3060 11275 \n", + "153 3160 9750 \n", + "... ... ... \n", + "21535 3990 7208 \n", + "21552 2190 10777 \n", + "21560 3470 18270 \n", + "21581 2290 10125 \n", + "21584 2780 8964 \n", + "\n", + "[805 rows x 20 columns]\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "556 5486800070 6/20/2014 1950000.0 7 3.50 4640 \n", + "1134 4024100951 1/5/2015 420000.0 7 3.00 2940 \n", + "1239 7227802030 6/23/2014 350000.0 7 3.00 2800 \n", + "1658 9126101740 12/4/2014 490000.0 8 5.00 2800 \n", + "3717 5451100490 1/15/2015 884900.0 7 4.75 5370 \n", + "... ... ... ... ... ... ... \n", + "18808 4040500100 10/20/2014 539000.0 7 2.25 2620 \n", + "18960 1778360150 6/20/2014 1240000.0 7 5.50 6630 \n", + "19239 8812401450 12/29/2014 660000.0 10 3.00 2920 \n", + "19287 3756900027 11/25/2014 575000.0 8 3.00 3840 \n", + "19312 2771604190 6/17/2014 824000.0 7 4.25 3670 \n", + "\n", + " sqft_lot floors view condition grade sqft_above \\\n", + "556 15235 2.0 FAIR Average 11 Excellent 2860 \n", + "1134 8624 1.0 NONE Average 8 Good 1690 \n", + "1239 9569 1.0 AVERAGE Average 7 Average 1400 \n", + "1658 2580 2.0 NONE Average 8 Good 1880 \n", + "3717 10800 1.5 NONE Average 8 Good 5370 \n", + "... ... ... ... ... ... ... \n", + "18808 6890 2.0 NONE Good 7 Average 2620 \n", + "18960 13782 2.0 NONE Average 10 Very Good 4930 \n", + "19239 3745 2.0 NONE Good 7 Average 1860 \n", + "19287 15990 1.0 NONE Average 7 Average 2530 \n", + "19312 4000 2.0 FAIR Average 8 Good 2800 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "556 1780.0 1965 2003.000000 98040 47.5666 -122.231 \n", + "1134 1250.0 1977 83.636778 98155 47.7555 -122.307 \n", + "1239 1400.0 1963 0.000000 98056 47.5102 -122.183 \n", + "1658 920.0 1997 0.000000 98122 47.6086 -122.303 \n", + "3717 0.0 1967 0.000000 98040 47.5380 -122.223 \n", + "... ... ... ... ... ... ... \n", + "18808 0.0 1961 0.000000 98007 47.6123 -122.134 \n", + "18960 1700.0 2004 83.636778 98006 47.5399 -122.118 \n", + "19239 1060.0 1913 0.000000 98105 47.6635 -122.320 \n", + "19287 1310.0 1961 0.000000 98034 47.7111 -122.211 \n", + "19312 870.0 1964 0.000000 98199 47.6375 -122.388 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "556 3230 20697 \n", + "1134 1850 8031 \n", + "1239 2150 7333 \n", + "1658 1800 2580 \n", + "3717 2310 10910 \n", + "... ... ... \n", + "18808 2070 7910 \n", + "18960 4470 8639 \n", + "19239 1810 3745 \n", + "19287 1380 8172 \n", + "19312 2010 4000 \n", + "\n", + "[62 rows x 20 columns]\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "5 7237550310 5/12/2014 1230000.0 4 4.50 5420 \n", + "270 4054500390 10/7/2014 1370000.0 4 4.75 5310 \n", + "300 3225069065 6/24/2014 3080000.0 4 5.00 4550 \n", + "419 8678500060 7/10/2014 1550000.0 5 4.25 6070 \n", + "450 4055700030 5/2/2015 1450000.0 3 4.50 3970 \n", + "... ... ... ... ... ... ... \n", + "21478 2413910120 7/2/2014 915000.0 3 4.50 3850 \n", + "21485 4233600150 2/3/2015 1150000.0 5 4.25 4010 \n", + "21490 2524069097 5/9/2014 2240000.0 5 6.50 7270 \n", + "21535 1561750040 12/24/2014 1380000.0 5 4.50 4350 \n", + "21560 9253900271 1/7/2015 3570000.0 5 4.50 4850 \n", + "\n", + " sqft_lot floors view condition grade sqft_above \\\n", + "5 101930 1.0 NONE Average 11 Excellent 3890 \n", + "270 57346 2.0 NONE Good 11 Excellent 5310 \n", + "300 18641 1.0 EXCELLENT Average 10 Very Good 2600 \n", + "419 171626 2.0 NONE Average 12 Luxury 6070 \n", + "450 24920 2.0 AVERAGE Average 10 Very Good 3260 \n", + "... ... ... ... ... ... ... \n", + "21478 62726 2.0 NONE Average 10 Very Good 3120 \n", + "21485 8252 2.0 NONE Average 10 Very Good 4010 \n", + "21490 130017 2.0 NONE Average 12 Luxury 6420 \n", + "21535 13405 2.0 NONE Average 11 Excellent 4350 \n", + "21560 10584 2.0 EXCELLENT Average 10 Very Good 3540 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "5 1530.0 2001 0.000000 98053 47.6561 -122.005 \n", + "270 0.0 1989 83.636778 98077 47.7285 -122.042 \n", + "300 1950.0 2002 0.000000 98074 47.6053 -122.077 \n", + "419 0.0 1999 0.000000 98024 47.5954 -121.950 \n", + "450 710.0 1977 83.636778 98034 47.7183 -122.258 \n", + "... ... ... ... ... ... ... \n", + "21478 730.0 2013 0.000000 98053 47.6735 -122.058 \n", + "21485 0.0 2015 83.636778 98075 47.5974 -122.013 \n", + "21490 850.0 2010 83.636778 98027 47.5371 -121.982 \n", + "21535 0.0 2014 0.000000 98074 47.6018 -122.060 \n", + "21560 1310.0 2007 0.000000 98008 47.5943 -122.110 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "5 4760 101930 \n", + "270 4180 47443 \n", + "300 4550 19508 \n", + "419 4680 211267 \n", + "450 2610 13838 \n", + "... ... ... \n", + "21478 2630 46609 \n", + "21485 3370 8252 \n", + "21490 1800 44890 \n", + "21535 3990 7208 \n", + "21560 3470 18270 \n", + "\n", + "[263 rows x 20 columns]\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "5 7237550310 5/12/2014 1230000.0 4 4.50 5420 \n", + "153 7855801670 4/1/2015 2250000.0 4 3.25 5180 \n", + "269 7960900060 5/4/2015 2900000.0 4 3.25 5050 \n", + "270 4054500390 10/7/2014 1370000.0 4 4.75 5310 \n", + "384 713500030 7/28/2014 1350000.0 5 3.50 4800 \n", + "... ... ... ... ... ... ... \n", + "21504 7237550100 8/25/2014 1410000.0 4 4.00 4920 \n", + "21505 7430500110 12/9/2014 1380000.0 5 3.50 5150 \n", + "21514 8964800330 4/7/2015 3000000.0 4 3.75 5090 \n", + "21560 9253900271 1/7/2015 3570000.0 5 4.50 4850 \n", + "21574 7430200100 5/14/2014 1220000.0 4 3.50 4910 \n", + "\n", + " sqft_lot floors view condition grade sqft_above \\\n", + "5 101930 1.0 NONE Average 11 Excellent 3890 \n", + "153 19850 2.0 GOOD Average 12 Luxury 3540 \n", + "269 20100 1.5 AVERAGE Average 11 Excellent 4750 \n", + "270 57346 2.0 NONE Good 11 Excellent 5310 \n", + "384 14984 2.0 AVERAGE Average 11 Excellent 3480 \n", + "... ... ... ... ... ... ... \n", + "21504 50621 2.0 NONE Average 10 Very Good 4280 \n", + "21505 12230 2.0 AVERAGE Average 10 Very Good 3700 \n", + "21514 14823 1.0 NONE Average 11 Excellent 4180 \n", + "21560 10584 2.0 EXCELLENT Average 10 Very Good 3540 \n", + "21574 9444 1.5 NONE Average 11 Excellent 3110 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "5 1530.0 2001 0.000000 98053 47.6561 -122.005 \n", + "153 1640.0 2006 0.000000 98006 47.5620 -122.162 \n", + "269 300.0 1982 83.636778 98004 47.6312 -122.223 \n", + "270 0.0 1989 83.636778 98077 47.7285 -122.042 \n", + "384 ? 1998 0.000000 98006 47.5543 -122.148 \n", + "... ... ... ... ... ... ... \n", + "21504 640.0 2012 0.000000 98053 47.6575 -122.006 \n", + "21505 1450.0 2007 0.000000 98008 47.6249 -122.090 \n", + "21514 910.0 2013 83.636778 98004 47.6200 -122.207 \n", + "21560 1310.0 2007 0.000000 98008 47.5943 -122.110 \n", + "21574 1800.0 2007 0.000000 98074 47.6502 -122.066 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "5 4760 101930 \n", + "153 3160 9750 \n", + "269 3890 20060 \n", + "270 4180 47443 \n", + "384 4050 19009 \n", + "... ... ... \n", + "21504 4920 74052 \n", + "21505 2940 13462 \n", + "21514 3030 12752 \n", + "21560 3470 18270 \n", + "21574 4560 11063 \n", + "\n", + "[255 rows x 20 columns]\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "5 7237550310 5/12/2014 1230000.0 4 4.50 5420 \n", + "21 2524049179 8/26/2014 2000000.0 3 2.75 3050 \n", + "36 9435300030 5/28/2014 550000.0 4 1.00 1660 \n", + "41 7766200013 8/11/2014 775000.0 4 2.25 4220 \n", + "49 822039084 3/11/2015 1350000.0 3 2.50 2753 \n", + "... ... ... ... ... ... ... \n", + "21478 2413910120 7/2/2014 915000.0 3 4.50 3850 \n", + "21490 2524069097 5/9/2014 2240000.0 5 6.50 7270 \n", + "21504 7237550100 8/25/2014 1410000.0 4 4.00 4920 \n", + "21509 2625069038 11/24/2014 1450000.0 4 3.50 4300 \n", + "21532 8835770330 8/19/2014 1060000.0 2 1.50 2370 \n", + "\n", + " sqft_lot floors view condition grade sqft_above \\\n", + "5 101930 1.0 NONE Average 11 Excellent 3890 \n", + "21 44867 1.0 EXCELLENT Average 9 Better 2330 \n", + "36 34848 1.0 NONE Poor 5 Fair 930 \n", + "41 24186 1.0 NONE Average 8 Good 2600 \n", + "49 65005 1.0 AVERAGE Very Good 9 Better 2165 \n", + "... ... ... ... ... ... ... \n", + "21478 62726 2.0 NONE Average 10 Very Good 3120 \n", + "21490 130017 2.0 NONE Average 12 Luxury 6420 \n", + "21504 50621 2.0 NONE Average 10 Very Good 4280 \n", + "21509 108865 2.0 NONE Average 11 Excellent 4300 \n", + "21532 184231 2.0 NONE Average 11 Excellent 2370 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "5 1530.0 2001 0.000000 98053 47.6561 -122.005 \n", + "21 720.0 1968 0.000000 98040 47.5316 -122.233 \n", + "36 730.0 1933 0.000000 98052 47.6621 -122.132 \n", + "41 1620.0 1984 0.000000 98166 47.4450 -122.347 \n", + "49 588.0 1953 0.000000 98070 47.4041 -122.451 \n", + "... ... ... ... ... ... ... \n", + "21478 730.0 2013 0.000000 98053 47.6735 -122.058 \n", + "21490 850.0 2010 83.636778 98027 47.5371 -121.982 \n", + "21504 640.0 2012 0.000000 98053 47.6575 -122.006 \n", + "21509 0.0 2014 0.000000 98074 47.6258 -122.005 \n", + "21532 0.0 2005 0.000000 98045 47.4543 -121.778 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "5 4760 101930 \n", + "21 4110 20336 \n", + "36 2160 11467 \n", + "41 2410 30617 \n", + "49 2680 72513 \n", + "... ... ... \n", + "21478 2630 46609 \n", + "21490 1800 44890 \n", + "21504 4920 74052 \n", + "21509 4650 107498 \n", + "21532 3860 151081 \n", + "\n", + "[2091 rows x 20 columns]\n", + "Empty DataFrame\n", + "Columns: [id, date, price, bedrooms, bathrooms, sqft_living, sqft_lot, floors, view, condition, grade, sqft_above, sqft_basement, yr_built, yr_renovated, zipcode, lat, long, sqft_living15, sqft_lot15]\n", + "Index: []\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "269 7960900060 5/4/2015 2900000.0 4 3.25 5050 \n", + "270 4054500390 10/7/2014 1370000.0 4 4.75 5310 \n", + "419 8678500060 7/10/2014 1550000.0 5 4.25 6070 \n", + "484 2625069070 4/10/2015 1390000.0 4 3.25 4860 \n", + "512 7322910030 7/21/2014 1100000.0 5 3.50 4410 \n", + "... ... ... ... ... ... ... \n", + "21509 2625069038 11/24/2014 1450000.0 4 3.50 4300 \n", + "21516 324069112 6/17/2014 1330000.0 4 4.00 4420 \n", + "21535 1561750040 12/24/2014 1380000.0 5 4.50 4350 \n", + "21545 6664500090 1/15/2015 750000.0 5 4.00 4500 \n", + "21584 249000205 10/15/2014 1540000.0 5 3.75 4470 \n", + "\n", + " sqft_lot floors view condition grade sqft_above \\\n", + "269 20100 1.5 AVERAGE Average 11 Excellent 4750 \n", + "270 57346 2.0 NONE Good 11 Excellent 5310 \n", + "419 171626 2.0 NONE Average 12 Luxury 6070 \n", + "484 181319 2.5 NONE Average 9 Better 4860 \n", + "512 57063 2.0 NONE Good 9 Better 4410 \n", + "... ... ... ... ... ... ... \n", + "21509 108865 2.0 NONE Average 11 Excellent 4300 \n", + "21516 16526 2.0 NONE Average 11 Excellent 4420 \n", + "21535 13405 2.0 NONE Average 11 Excellent 4350 \n", + "21545 8130 2.0 NONE Average 10 Very Good 4500 \n", + "21584 8088 2.0 NONE Average 11 Excellent 4470 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "269 300.0 1982 83.636778 98004 47.6312 -122.223 \n", + "270 0.0 1989 83.636778 98077 47.7285 -122.042 \n", + "419 0.0 1999 0.000000 98024 47.5954 -121.950 \n", + "484 0.0 1993 0.000000 98074 47.6179 -122.005 \n", + "512 0.0 1990 0.000000 98053 47.6554 -122.018 \n", + "... ... ... ... ... ... ... \n", + "21509 0.0 2014 0.000000 98074 47.6258 -122.005 \n", + "21516 0.0 2013 0.000000 98075 47.5914 -122.027 \n", + "21535 0.0 2014 0.000000 98074 47.6018 -122.060 \n", + "21545 0.0 2007 0.000000 98059 47.4832 -122.145 \n", + "21584 0.0 2008 0.000000 98004 47.6321 -122.200 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "269 3890 20060 \n", + "270 4180 47443 \n", + "419 4680 211267 \n", + "484 3850 181319 \n", + "512 2900 50529 \n", + "... ... ... \n", + "21509 4650 107498 \n", + "21516 3510 50447 \n", + "21535 3990 7208 \n", + "21545 2840 8402 \n", + "21584 2780 8964 \n", + "\n", + "[264 rows x 20 columns]\n", + "Empty DataFrame\n", + "Columns: [id, date, price, bedrooms, bathrooms, sqft_living, sqft_lot, floors, view, condition, grade, sqft_above, sqft_basement, yr_built, yr_renovated, zipcode, lat, long, sqft_living15, sqft_lot15]\n", + "Index: []\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "1 6414100192 12/9/2014 538000.0 3 2.25 2570 \n", + "2 5631500400 2/25/2015 180000.0 2 1.00 770 \n", + "12 114101516 5/28/2014 310000.0 3 1.00 1430 \n", + "23 8091400200 5/16/2014 252700.0 2 1.50 1070 \n", + "26 1794500383 6/26/2014 937000.0 3 1.75 2450 \n", + "... ... ... ... ... ... ... \n", + "21576 1931300412 4/16/2015 475000.0 3 2.25 1190 \n", + "21577 8672200110 3/17/2015 1090000.0 5 3.75 4170 \n", + "21579 1972201967 10/31/2014 520000.0 2 2.25 1530 \n", + "21581 191100405 4/21/2015 1580000.0 4 3.25 3410 \n", + "21583 7202300110 9/15/2014 810000.0 4 3.00 3990 \n", + "\n", + " sqft_lot floors view condition grade sqft_above \\\n", + "1 7242 2.0 NONE Average 7 Average 2170 \n", + "2 10000 1.0 NONE Average 6 Low Average 770 \n", + "12 19901 1.5 NONE Good 7 Average 1430 \n", + "23 9643 1.0 NONE Average 7 Average 1070 \n", + "26 2691 2.0 NONE Average 8 Good 1750 \n", + "... ... ... ... ... ... ... \n", + "21576 1200 3.0 NONE Average 8 Good 1190 \n", + "21577 8142 2.0 AVERAGE Average 10 Very Good 4170 \n", + "21579 981 3.0 NONE Average 8 Good 1480 \n", + "21581 10125 2.0 NONE Average 10 Very Good 3410 \n", + "21583 7838 2.0 NONE Average 9 Better 3990 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "1 400.0 1951 1991.000000 98125 47.7210 -122.319 \n", + "2 0.0 1933 83.636778 98028 47.7379 -122.233 \n", + "12 0.0 1927 83.636778 98028 47.7558 -122.229 \n", + "23 0.0 1985 83.636778 98030 47.3533 -122.166 \n", + "26 700.0 1915 83.636778 98119 47.6386 -122.360 \n", + "... ... ... ... ... ... ... \n", + "21576 0.0 2008 83.636778 98103 47.6542 -122.346 \n", + "21577 0.0 2006 83.636778 98056 47.5354 -122.181 \n", + "21579 50.0 2006 83.636778 98103 47.6533 -122.346 \n", + "21581 ? 2007 83.636778 98040 47.5653 -122.223 \n", + "21583 0.0 2003 83.636778 98053 47.6857 -122.046 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "1 1690 7639 \n", + "2 2720 8062 \n", + "12 1780 12697 \n", + "23 1220 8386 \n", + "26 1760 3573 \n", + "... ... ... \n", + "21576 1180 1224 \n", + "21577 3030 7980 \n", + "21579 1530 1282 \n", + "21581 2290 10125 \n", + "21583 3370 6814 \n", + "\n", + "[4573 rows x 20 columns]\n", + "Empty DataFrame\n", + "Columns: [id, date, price, bedrooms, bathrooms, sqft_living, sqft_lot, floors, view, condition, grade, sqft_above, sqft_basement, yr_built, yr_renovated, zipcode, lat, long, sqft_living15, sqft_lot15]\n", + "Index: []\n", + "Empty DataFrame\n", + "Columns: [id, date, price, bedrooms, bathrooms, sqft_living, sqft_lot, floors, view, condition, grade, sqft_above, sqft_basement, yr_built, yr_renovated, zipcode, lat, long, sqft_living15, sqft_lot15]\n", + "Index: []\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "99 7338200240 5/16/2014 437500.0 3 2.50 2320 \n", + "252 1422300030 4/1/2015 415000.0 3 2.25 1510 \n", + "365 723099065 1/30/2015 465000.0 3 2.00 1840 \n", + "1610 1422300140 9/5/2014 454000.0 3 2.50 2530 \n", + "2587 5061300030 5/8/2014 134000.0 2 1.50 980 \n", + "2608 1422300160 6/24/2014 379000.0 3 2.50 1740 \n", + "2845 1723099031 10/20/2014 724950.0 4 3.50 3010 \n", + "2925 8649401270 4/30/2015 167000.0 1 1.00 780 \n", + "3087 1823099076 8/20/2014 495000.0 3 2.50 1780 \n", + "3292 9413400165 6/24/2014 380000.0 3 2.25 1860 \n", + "3320 1323089184 5/2/2014 452500.0 3 2.50 2430 \n", + "4199 1437500015 7/9/2014 150000.0 3 0.75 490 \n", + "4844 192300020 5/21/2014 525000.0 3 2.75 2100 \n", + "5861 7349800780 8/5/2014 175000.0 2 1.75 1050 \n", + "6083 5062300280 4/16/2015 150000.0 3 1.00 890 \n", + "6464 723099028 6/26/2014 320000.0 3 2.00 1550 \n", + "6721 7338200170 4/22/2015 600000.0 4 2.50 2710 \n", + "8104 1422300100 9/29/2014 435000.0 3 2.50 1730 \n", + "8290 1923099058 10/15/2014 620000.0 4 2.50 2980 \n", + "9085 1223089081 5/30/2014 425000.0 3 1.75 1510 \n", + "9425 1223089038 7/11/2014 665000.0 5 2.25 3320 \n", + "10084 7805600070 11/11/2014 200000.0 2 1.75 1320 \n", + "10886 8649401000 10/22/2014 241000.0 2 1.75 1070 \n", + "12808 723099044 8/7/2014 433200.0 3 2.50 2075 \n", + "12910 7338200180 9/10/2014 590000.0 4 2.50 2660 \n", + "13059 1437500035 10/10/2014 155000.0 2 1.00 1010 \n", + "13236 8649400410 4/17/2015 375000.0 3 1.75 2140 \n", + "13981 774100355 11/3/2014 370000.0 2 2.00 2100 \n", + "14601 8649400790 1/13/2015 160000.0 3 1.00 1340 \n", + "14722 4032500035 9/13/2014 295000.0 2 1.75 1560 \n", + "15920 1823099056 12/22/2014 745000.0 3 2.50 2810 \n", + "15995 1923099034 1/16/2015 775000.0 4 3.50 3970 \n", + "16828 226109056 3/26/2015 170000.0 1 0.75 850 \n", + "16927 2626119028 3/23/2015 160000.0 3 1.00 1140 \n", + "17379 1823099028 7/22/2014 440000.0 3 2.00 1790 \n", + "19634 2626119062 11/12/2014 155000.0 3 1.00 1300 \n", + "19964 774100475 6/27/2014 415000.0 3 2.75 2600 \n", + "21370 774101755 4/17/2015 320000.0 3 1.75 1790 \n", + "\n", + " sqft_lot floors view condition grade sqft_above \\\n", + "99 36847 2.0 AVERAGE Average 9 Better 2320 \n", + "252 36224 2.0 NONE Average 8 Good 1510 \n", + "365 40438 2.0 NONE Average 7 Average 1840 \n", + "1610 43733 2.0 NONE Average 8 Good 1530 \n", + "2587 5000 2.0 NONE Average 7 Average 980 \n", + "2608 30886 2.0 NONE Average 8 Good 1740 \n", + "2845 174240 2.0 NONE Average 9 Better 3010 \n", + "2925 10235 1.5 NONE Average 6 Low Average 780 \n", + "3087 47480 2.0 NONE Average 7 Average 1780 \n", + "3292 15559 2.0 NONE Good 7 Average 1860 \n", + "3320 88426 1.0 NONE Good 7 Average 1570 \n", + "4199 38500 1.5 NONE Good 5 Fair 490 \n", + "4844 10362 2.0 NONE Average 9 Better 1510 \n", + "5861 9800 1.5 NONE Good 6 Low Average 1050 \n", + "6083 6488 1.5 NONE Average 5 Fair 890 \n", + "6464 34175 1.5 NONE Average 7 Average 1550 \n", + "6721 35009 2.0 AVERAGE Average 9 Better 2710 \n", + "8104 46638 2.0 NONE Average 8 Good 1730 \n", + "8290 210395 2.0 NONE Average 9 Better 2980 \n", + "9085 44000 1.0 NONE Average 7 Average 1240 \n", + "9425 60984 2.0 NONE Average 9 Better 3320 \n", + "10084 13052 1.5 NONE Average 7 Average 1320 \n", + "10886 9750 1.5 NONE Average 7 Average 1070 \n", + "12808 16200 2.0 NONE Average 8 Good 2075 \n", + "12910 35010 2.0 AVERAGE Average 9 Better 2660 \n", + "13059 43056 1.5 NONE Average 5 Fair 1010 \n", + "13236 13598 1.5 NONE Good 7 Average 1620 \n", + "13981 58488 2.0 NONE Average 9 Better 2100 \n", + "14601 18552 1.5 NONE Good 5 Fair 1340 \n", + "14722 43748 2.0 NONE Average 8 Good 1560 \n", + "15920 435600 2.0 NONE Average 9 Better 2810 \n", + "15995 210830 2.0 NONE Average 9 Better 3970 \n", + "16828 5600 1.0 AVERAGE Average 6 Low Average 850 \n", + "16927 3240 1.5 NONE Good 6 Low Average 1140 \n", + "17379 32379 1.0 NONE Average 7 Average 1790 \n", + "19634 6098 1.0 NONE Average 7 Average 1300 \n", + "19964 64626 1.5 NONE Average 8 Good 2600 \n", + "21370 66250 1.5 NONE Average 7 Average 1790 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "99 0.0 1992 0.000000 98045 47.4838 -121.714 \n", + "252 0.0 1991 0.000000 98045 47.4616 -121.711 \n", + "365 0.0 1994 83.636778 98045 47.4853 -121.709 \n", + "1610 1000.0 1991 0.000000 98045 47.4600 -121.708 \n", + "2587 0.0 1922 2003.000000 98014 47.7076 -121.359 \n", + "2608 0.0 1992 0.000000 98045 47.4600 -121.707 \n", + "2845 0.0 2004 0.000000 98045 47.4775 -121.691 \n", + "2925 0.0 1989 83.636778 98014 47.7130 -121.315 \n", + "3087 0.0 1995 83.636778 98045 47.4723 -121.707 \n", + "3292 0.0 1963 0.000000 98022 47.1559 -121.646 \n", + "3320 860.0 1985 0.000000 98045 47.4828 -121.718 \n", + "4199 0.0 1959 0.000000 98014 47.7112 -121.315 \n", + "4844 590.0 1998 0.000000 98045 47.4347 -121.417 \n", + "5861 0.0 1975 0.000000 98019 47.7595 -121.473 \n", + "6083 0.0 1928 0.000000 98014 47.7087 -121.352 \n", + "6464 0.0 1999 0.000000 98045 47.4855 -121.698 \n", + "6721 0.0 1992 0.000000 98045 47.4815 -121.714 \n", + "8104 0.0 1991 0.000000 98045 47.4614 -121.709 \n", + "8290 0.0 2001 0.000000 98045 47.4575 -121.707 \n", + "9085 270.0 1989 0.000000 98045 47.4851 -121.716 \n", + "9425 0.0 2000 0.000000 98045 47.4862 -121.718 \n", + "10084 0.0 1980 0.000000 98014 47.7120 -121.352 \n", + "10886 0.0 1995 0.000000 98014 47.7131 -121.319 \n", + "12808 ? 2002 0.000000 98045 47.4848 -121.698 \n", + "12910 0.0 1993 83.636778 98045 47.4816 -121.714 \n", + "13059 0.0 1990 83.636778 98014 47.7105 -121.316 \n", + "13236 520.0 1970 0.000000 98014 47.7139 -121.321 \n", + "13981 0.0 2005 83.636778 98014 47.7200 -121.402 \n", + "14601 0.0 1935 0.000000 98014 47.7129 -121.325 \n", + "14722 0.0 1967 2000.000000 98065 47.5729 -121.676 \n", + "15920 0.0 1995 0.000000 98045 47.4816 -121.701 \n", + "15995 0.0 2000 0.000000 98045 47.4614 -121.713 \n", + "16828 0.0 1903 83.636778 98019 47.7654 -121.480 \n", + "16927 0.0 1910 0.000000 98014 47.7093 -121.364 \n", + "17379 0.0 2007 0.000000 98045 47.4826 -121.698 \n", + "19634 ? 2013 0.000000 98014 47.7074 -121.364 \n", + "19964 0.0 2009 0.000000 98014 47.7185 -121.405 \n", + "21370 0.0 2003 0.000000 98014 47.7179 -121.403 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "99 2550 35065 \n", + "252 1730 36224 \n", + "365 1380 44049 \n", + "1610 1730 43548 \n", + "2587 1040 5000 \n", + "2608 1740 39133 \n", + "2845 2720 247856 \n", + "2925 930 10165 \n", + "3087 1890 51836 \n", + "3292 1110 11586 \n", + "3320 1560 56827 \n", + "4199 800 18297 \n", + "4844 2240 11842 \n", + "5861 1230 12726 \n", + "6083 1330 16250 \n", + "6464 2300 35174 \n", + "6721 2330 35040 \n", + "8104 1750 35508 \n", + "8290 2530 45596 \n", + "9085 2290 36242 \n", + "9425 1580 55322 \n", + "10084 1320 13052 \n", + "10886 970 9750 \n", + "12808 2300 32379 \n", + "12910 2330 35448 \n", + "13059 830 18297 \n", + "13236 930 10150 \n", + "13981 1440 59346 \n", + "14601 960 15141 \n", + "14722 1000 24602 \n", + "15920 2380 92007 \n", + "15995 1680 42665 \n", + "16828 900 12250 \n", + "16927 1140 4700 \n", + "17379 2290 43560 \n", + "19634 1300 6849 \n", + "19964 1740 64626 \n", + "21370 1440 59346 \n", + " id date price bedrooms bathrooms sqft_living \\\n", + "5 7237550310 5/12/2014 1230000.0 4 4.50 5420 \n", + "21 2524049179 8/26/2014 2000000.0 3 2.75 3050 \n", + "270 4054500390 10/7/2014 1370000.0 4 4.75 5310 \n", + "300 3225069065 6/24/2014 3080000.0 4 5.00 4550 \n", + "419 8678500060 7/10/2014 1550000.0 5 4.25 6070 \n", + "... ... ... ... ... ... ... \n", + "21470 98300230 4/28/2015 1460000.0 4 4.00 4620 \n", + "21504 7237550100 8/25/2014 1410000.0 4 4.00 4920 \n", + "21509 2625069038 11/24/2014 1450000.0 4 3.50 4300 \n", + "21524 715010530 1/13/2015 1880000.0 5 3.50 4410 \n", + "21574 7430200100 5/14/2014 1220000.0 4 3.50 4910 \n", + "\n", + " sqft_lot floors view condition grade sqft_above \\\n", + "5 101930 1.0 NONE Average 11 Excellent 3890 \n", + "21 44867 1.0 EXCELLENT Average 9 Better 2330 \n", + "270 57346 2.0 NONE Good 11 Excellent 5310 \n", + "300 18641 1.0 EXCELLENT Average 10 Very Good 2600 \n", + "419 171626 2.0 NONE Average 12 Luxury 6070 \n", + "... ... ... ... ... ... ... \n", + "21470 130208 2.0 NONE Average 10 Very Good 4620 \n", + "21504 50621 2.0 NONE Average 10 Very Good 4280 \n", + "21509 108865 2.0 NONE Average 11 Excellent 4300 \n", + "21524 13000 2.0 GOOD Average 10 Very Good 2920 \n", + "21574 9444 1.5 NONE Average 11 Excellent 3110 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "5 1530.0 2001 0.000000 98053 47.6561 -122.005 \n", + "21 720.0 1968 0.000000 98040 47.5316 -122.233 \n", + "270 0.0 1989 83.636778 98077 47.7285 -122.042 \n", + "300 1950.0 2002 0.000000 98074 47.6053 -122.077 \n", + "419 0.0 1999 0.000000 98024 47.5954 -121.950 \n", + "... ... ... ... ... ... ... \n", + "21470 0.0 2014 0.000000 98024 47.5885 -121.939 \n", + "21504 640.0 2012 0.000000 98053 47.6575 -122.006 \n", + "21509 0.0 2014 0.000000 98074 47.6258 -122.005 \n", + "21524 1490.0 2014 0.000000 98006 47.5382 -122.111 \n", + "21574 1800.0 2007 0.000000 98074 47.6502 -122.066 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "5 4760 101930 \n", + "21 4110 20336 \n", + "270 4180 47443 \n", + "300 4550 19508 \n", + "419 4680 211267 \n", + "... ... ... \n", + "21470 4620 131007 \n", + "21504 4920 74052 \n", + "21509 4650 107498 \n", + "21524 5790 12969 \n", + "21574 4560 11063 \n", + "\n", + "[200 rows x 20 columns]\n", + " id date price bedrooms bathrooms sqft_living \\\n", + "5 7237550310 5/12/2014 1230000.0 4 4.50 5420 \n", + "21 2524049179 8/26/2014 2000000.0 3 2.75 3050 \n", + "41 7766200013 8/11/2014 775000.0 4 2.25 4220 \n", + "49 822039084 3/11/2015 1350000.0 3 2.50 2753 \n", + "69 1802000060 6/12/2014 1330000.0 5 2.25 3200 \n", + "... ... ... ... ... ... ... \n", + "21490 2524069097 5/9/2014 2240000.0 5 6.50 7270 \n", + "21504 7237550100 8/25/2014 1410000.0 4 4.00 4920 \n", + "21509 2625069038 11/24/2014 1450000.0 4 3.50 4300 \n", + "21516 324069112 6/17/2014 1330000.0 4 4.00 4420 \n", + "21532 8835770330 8/19/2014 1060000.0 2 1.50 2370 \n", + "\n", + " sqft_lot floors view condition grade sqft_above \\\n", + "5 101930 1.0 NONE Average 11 Excellent 3890 \n", + "21 44867 1.0 EXCELLENT Average 9 Better 2330 \n", + "41 24186 1.0 NONE Average 8 Good 2600 \n", + "49 65005 1.0 AVERAGE Very Good 9 Better 2165 \n", + "69 20158 1.0 NONE Average 8 Good 1600 \n", + "... ... ... ... ... ... ... \n", + "21490 130017 2.0 NONE Average 12 Luxury 6420 \n", + "21504 50621 2.0 NONE Average 10 Very Good 4280 \n", + "21509 108865 2.0 NONE Average 11 Excellent 4300 \n", + "21516 16526 2.0 NONE Average 11 Excellent 4420 \n", + "21532 184231 2.0 NONE Average 11 Excellent 2370 \n", + "\n", + " sqft_basement yr_built yr_renovated zipcode lat long \\\n", + "5 1530.0 2001 0.000000 98053 47.6561 -122.005 \n", + "21 720.0 1968 0.000000 98040 47.5316 -122.233 \n", + "41 1620.0 1984 0.000000 98166 47.4450 -122.347 \n", + "49 588.0 1953 0.000000 98070 47.4041 -122.451 \n", + "69 1600.0 1965 0.000000 98004 47.6303 -122.215 \n", + "... ... ... ... ... ... ... \n", + "21490 850.0 2010 83.636778 98027 47.5371 -121.982 \n", + "21504 640.0 2012 0.000000 98053 47.6575 -122.006 \n", + "21509 0.0 2014 0.000000 98074 47.6258 -122.005 \n", + "21516 0.0 2013 0.000000 98075 47.5914 -122.027 \n", + "21532 0.0 2005 0.000000 98045 47.4543 -121.778 \n", + "\n", + " sqft_living15 sqft_lot15 \n", + "5 4760 101930 \n", + "21 4110 20336 \n", + "41 2410 30617 \n", + "49 2680 72513 \n", + "69 3390 20158 \n", + "... ... ... \n", + "21490 1800 44890 \n", + "21504 4920 74052 \n", + "21509 4650 107498 \n", + "21516 3510 50447 \n", + "21532 3860 151081 \n", + "\n", + "[1897 rows x 20 columns]\n" + ] + } + ], + "source": [ + "def detect_outliers_iqr(df, threshold=1.5):\n", + " outliers = {}\n", + " for column in df.columns:\n", + " if df[column].dtype in ['int64', 'float64']:\n", + " Q1 = df[column].quantile(0.25)\n", + " Q3 = df[column].quantile(0.75)\n", + " IQR = Q3 - Q1\n", + " lower_bound = Q1 - threshold * IQR\n", + " upper_bound = Q3 + threshold * IQR\n", + " outlier_indices = df[(df[column] < lower_bound) | (df[column] > upper_bound)].index\n", + " outliers[column] = df.loc[outlier_indices]\n", + " return outliers\n", + "\n", + "\n", + "outliers_dict = detect_outliers_iqr(house_data, threshold=2.0)\n", + "\n", + "# Print outliers for each column\n", + "for column, outliers_df in outliers_dict.items():\n", + " print(outliers_df)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Replacing the outliers with the mean" + ] + }, + { + "cell_type": "code", + "execution_count": 288, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Index: 21534 entries, 0 to 21596\n", + "Data columns (total 20 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 21534 non-null float64\n", + " 1 date 21534 non-null object \n", + " 2 price 21534 non-null float64\n", + " 3 bedrooms 21534 non-null float64\n", + " 4 bathrooms 21534 non-null float64\n", + " 5 sqft_living 21534 non-null float64\n", + " 6 sqft_lot 21534 non-null float64\n", + " 7 floors 21534 non-null float64\n", + " 8 view 21534 non-null object \n", + " 9 condition 21534 non-null object \n", + " 10 grade 21534 non-null object \n", + " 11 sqft_above 21534 non-null float64\n", + " 12 sqft_basement 21534 non-null object \n", + " 13 yr_built 21534 non-null float64\n", + " 14 yr_renovated 21534 non-null float64\n", + " 15 zipcode 21534 non-null float64\n", + " 16 lat 21534 non-null float64\n", + " 17 long 21534 non-null float64\n", + " 18 sqft_living15 21534 non-null float64\n", + " 19 sqft_lot15 21534 non-null float64\n", + "dtypes: float64(15), object(5)\n", + "memory usage: 4.0+ MB\n", + "None\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_1243693/2423783818.py:12: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '4582351016.287034' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n", + " df.loc[outlier_indices, column] = df[column].mean()\n", + "/tmp/ipykernel_1243693/2423783818.py:12: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '3.3730379864400484' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n", + " df.loc[outlier_indices, column] = df[column].mean()\n", + "/tmp/ipykernel_1243693/2423783818.py:12: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '2079.8278536268226' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n", + " df.loc[outlier_indices, column] = df[column].mean()\n", + "/tmp/ipykernel_1243693/2423783818.py:12: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '15090.596359245843' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n", + " df.loc[outlier_indices, column] = df[column].mean()\n", + "/tmp/ipykernel_1243693/2423783818.py:12: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '1788.5575369183616' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n", + " df.loc[outlier_indices, column] = df[column].mean()\n", + "/tmp/ipykernel_1243693/2423783818.py:12: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '1971.0022754713477' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n", + " df.loc[outlier_indices, column] = df[column].mean()\n", + "/tmp/ipykernel_1243693/2423783818.py:12: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '98077.93935172286' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n", + " df.loc[outlier_indices, column] = df[column].mean()\n", + "/tmp/ipykernel_1243693/2423783818.py:12: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '1986.2999442741711' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n", + " df.loc[outlier_indices, column] = df[column].mean()\n", + "/tmp/ipykernel_1243693/2423783818.py:12: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '12751.079502182594' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n", + " df.loc[outlier_indices, column] = df[column].mean()\n" + ] + } + ], + "source": [ + "def replace_outliers_with_mean(df, threshold=1.5):\n", + " for column in df.columns:\n", + " if df[column].dtype in ['int64', 'float64']:\n", + " Q1 = df[column].quantile(0.25)\n", + " Q3 = df[column].quantile(0.75)\n", + " IQR = Q3 - Q1\n", + " lower_bound = Q1 - threshold * IQR\n", + " upper_bound = Q3 + threshold * IQR\n", + " \n", + " # Replace outliers with the mean\n", + " outlier_indices = (df[column] < lower_bound) | (df[column] > upper_bound)\n", + " df.loc[outlier_indices, column] = df[column].mean()\n", + " return df\n", + "\n", + "house_data_cleaned = replace_outliers_with_mean(house_data, threshold=1.5)\n", + "print(house_data.info())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Data Understanding" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import matplotlib.pyplot as plt\n" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": {}, + "outputs": [], + "source": [ + "# Load the dataset into a pandas DataFrame\n", + "house_data = pd.read_csv('data/kc_house_data.csv')\n" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "RangeIndex: 21597 entries, 0 to 21596\n", + "Data columns (total 21 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 id 21597 non-null int64 \n", + " 1 date 21597 non-null object \n", + " 2 price 21597 non-null float64\n", + " 3 bedrooms 21597 non-null int64 \n", + " 4 bathrooms 21597 non-null float64\n", + " 5 sqft_living 21597 non-null int64 \n", + " 6 sqft_lot 21597 non-null int64 \n", + " 7 floors 21597 non-null float64\n", + " 8 waterfront 19221 non-null object \n", + " 9 view 21534 non-null object \n", + " 10 condition 21597 non-null object \n", + " 11 grade 21597 non-null object \n", + " 12 sqft_above 21597 non-null int64 \n", + " 13 sqft_basement 21597 non-null object \n", + " 14 yr_built 21597 non-null int64 \n", + " 15 yr_renovated 17755 non-null float64\n", + " 16 zipcode 21597 non-null int64 \n", + " 17 lat 21597 non-null float64\n", + " 18 long 21597 non-null float64\n", + " 19 sqft_living15 21597 non-null int64 \n", + " 20 sqft_lot15 21597 non-null int64 \n", + "dtypes: float64(6), int64(9), object(6)\n", + "memory usage: 3.5+ MB\n" + ] + } + ], + "source": [ + "# Info of Data\n", + "inf = house_data.info()\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# EDA" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "First we further refine the dataset to make it more suitable for visualization. Unused columns are dropped and price formatted to thousands." + ] + }, + { + "cell_type": "code", + "execution_count": 289, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
pricebedroomsbathroomssqft_livingsqft_lotfloorsviewconditiongradesqft_abovesqft_basementyr_builtyr_renovatedzipcodelatlongsqft_living15sqft_lot15
0221.93.01.001180.05650.01.0NONEAverage7 Average1180.00.01955.00.00000098178.047.5112-122.2571340.05650.0
1538.03.02.252570.07242.02.0NONEAverage7 Average2170.0400.01951.083.74221598125.047.7210-122.3191690.07639.0
2180.02.01.00770.010000.01.0NONEAverage6 Low Average770.00.01933.083.74221598028.047.7379-122.2332720.08062.0
3604.04.03.001960.05000.01.0NONEVery Good7 Average1050.0910.01965.00.00000098136.047.5208-122.3931360.05000.0
4510.03.02.001680.08080.01.0NONEAverage8 Good1680.00.01987.00.00000098074.047.6168-122.0451800.07503.0
\n", + "
" + ], + "text/plain": [ + " price bedrooms bathrooms sqft_living sqft_lot floors view condition \\\n", + "0 221.9 3.0 1.00 1180.0 5650.0 1.0 NONE Average \n", + "1 538.0 3.0 2.25 2570.0 7242.0 2.0 NONE Average \n", + "2 180.0 2.0 1.00 770.0 10000.0 1.0 NONE Average \n", + "3 604.0 4.0 3.00 1960.0 5000.0 1.0 NONE Very Good \n", + "4 510.0 3.0 2.00 1680.0 8080.0 1.0 NONE Average \n", + "\n", + " grade sqft_above sqft_basement yr_built yr_renovated zipcode \\\n", + "0 7 Average 1180.0 0.0 1955.0 0.000000 98178.0 \n", + "1 7 Average 2170.0 400.0 1951.0 83.742215 98125.0 \n", + "2 6 Low Average 770.0 0.0 1933.0 83.742215 98028.0 \n", + "3 7 Average 1050.0 910.0 1965.0 0.000000 98136.0 \n", + "4 8 Good 1680.0 0.0 1987.0 0.000000 98074.0 \n", + "\n", + " lat long sqft_living15 sqft_lot15 \n", + "0 47.5112 -122.257 1340.0 5650.0 \n", + "1 47.7210 -122.319 1690.0 7639.0 \n", + "2 47.7379 -122.233 2720.0 8062.0 \n", + "3 47.5208 -122.393 1360.0 5000.0 \n", + "4 47.6168 -122.045 1800.0 7503.0 " + ] + }, + "execution_count": 289, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "eda_df = house_data_cleaned.copy()\n", + "eda_df.drop(\"date\", axis=1, inplace=True)\n", + "eda_df.drop(\"id\", axis=1, inplace=True)\n", + "eda_df.price = eda_df.price / 1000\n", + "eda_df.sqft_basement = eda_df.sqft_basement.apply(lambda x : 0 if x == \"?\" else x)\n", + "eda_df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Then we get a breakdown of the data in the price column." + ] + }, + { + "cell_type": "code", + "execution_count": 290, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "count 21534.000000\n", + "mean 480.083506\n", + "std 202.673668\n", + "min 78.000000\n", + "25% 322.000000\n", + "50% 450.000000\n", + "75% 590.000000\n", + "max 1120.000000\n", + "Name: price, dtype: float64" + ] + }, + "execution_count": 290, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "eda_df.price.describe()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Then we reduce the grade column to only the number rating for visualization." + ] + }, + { + "cell_type": "code", + "execution_count": 291, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "grade\n", + "7 8948\n", + "8 6053\n", + "9 2604\n", + "6 2031\n", + "10 1130\n", + "11 397\n", + "5 242\n", + "12 88\n", + "4 27\n", + "13 13\n", + "3 1\n", + "Name: count, dtype: int64" + ] + }, + "execution_count": 291, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "eda_df.grade = eda_df.grade.apply(lambda x : x.split(\" \")[0].strip())\n", + "eda_df.grade.value_counts()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Some functions were created to convert categorical columns to a numerical data type." + ] + }, + { + "cell_type": "code", + "execution_count": 292, + "metadata": {}, + "outputs": [], + "source": [ + "def column_values_list(column):\n", + " return list(dict(eda_df[column].value_counts().items()))\n", + "\n", + "def str_column_to_int(column):\n", + " replacement = 0\n", + " for value in column_values_list(column):\n", + " eda_df[column] = eda_df[column].apply(lambda x : replacement if x == value else x)\n", + " replacement += 1" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We then change the categorical column \"view\" to a numerical one." + ] + }, + { + "cell_type": "code", + "execution_count": 293, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "view\n", + "NONE 19422\n", + "AVERAGE 957\n", + "GOOD 508\n", + "FAIR 330\n", + "EXCELLENT 317\n", + "Name: count, dtype: int64" + ] + }, + "execution_count": 293, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "eda_df.view.value_counts()" + ] + }, + { + "cell_type": "code", + "execution_count": 294, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "view\n", + "0 19422\n", + "1 957\n", + "2 508\n", + "3 330\n", + "4 317\n", + "Name: count, dtype: int64" + ] + }, + "execution_count": 294, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "str_column_to_int(\"view\")\n", + "eda_df.view.value_counts()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Categorical column \"condition\" is also changed to a numeric value." + ] + }, + { + "cell_type": "code", + "execution_count": 295, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "condition\n", + "0 13981\n", + "1 5657\n", + "2 1697\n", + "3 170\n", + "4 29\n", + "Name: count, dtype: int64" + ] + }, + "execution_count": 295, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "str_column_to_int(\"condition\")\n", + "eda_df.condition.value_counts()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We then view the correlation of price against various columns." + ] + }, + { + "cell_type": "code", + "execution_count": 296, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "price 1.000000\n", + "sqft_living 0.581669\n", + "grade 0.573017\n", + "sqft_living15 0.508827\n", + "sqft_above 0.474044\n", + "lat 0.432947\n", + "bathrooms 0.424381\n", + "bedrooms 0.290729\n", + "floors 0.274164\n", + "sqft_basement 0.222340\n", + "view 0.203565\n", + "long 0.092386\n", + "sqft_lot 0.086452\n", + "sqft_lot15 0.067868\n", + "yr_built 0.061169\n", + "yr_renovated 0.030432\n", + "condition -0.013279\n", + "zipcode -0.021363\n", + "Name: price, dtype: float64" + ] + }, + "execution_count": 296, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "eda_df.corr()[\"price\"].sort_values(ascending=False)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Observations\n", + "\n", + "It seems that living area square footage is the most correlated with the price of the house. Conversely, the zipcode and condition of the house are negatively correlated with the price. This indicates that asa they increase, the price of the house decreases." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Visualizations" + ] + }, + { + "cell_type": "code", + "execution_count": 297, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import seaborn as sns\n", + "\n", + "sns.set_style(\"darkgrid\")\n", + "\n", + "corr_matrix = eda_df.corr().sort_values(\"price\")\n", + "sns.heatmap(corr_matrix);" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Observations\n", + "\n", + "The heat map confirms the correlation table we saw before." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "A function was created to visualize price against various columns." + ] + }, + { + "cell_type": "code", + "execution_count": 298, + "metadata": {}, + "outputs": [], + "source": [ + "def price_vs_column(column, alias, reg = True):\n", + " if reg:\n", + " ax = sns.regplot(\n", + " x=eda_df[column],\n", + " y=eda_df.price,\n", + " line_kws={\n", + " \"color\" : \"green\"\n", + " }\n", + " )\n", + " ax.set(\n", + " xlabel=alias,\n", + " ylabel=\"Price (in Thousands)\",\n", + " title=\"{} vs Price\".format(alias)\n", + " );\n", + " else:\n", + " ax = sns.scatterplot(\n", + " x=eda_df[column],\n", + " y=eda_df.price,\n", + " )\n", + " ax.set(\n", + " xlabel=alias,\n", + " ylabel=\"Price (in Thousands)\",\n", + " title=\"{} vs Price\".format(alias)\n", + " );" + ] + }, + { + "cell_type": "code", + "execution_count": 299, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "price_vs_column(\"sqft_living\", \"Living Area Sq. Ft\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Observations\n", + "\n", + "We see that as living area square footage increases, so does the price." + ] + }, + { + "cell_type": "code", + "execution_count": 300, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "price_vs_column(\"grade\", \"Construction/Design Grade\", reg=False)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Observations\n", + "\n", + "All grades contain prices across the whole spectrum." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Next, we view the effect of number of bedrooms on price." + ] + }, + { + "cell_type": "code", + "execution_count": 301, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "bedrooms\n", + "3.000000 9791\n", + "4.000000 6865\n", + "2.000000 2754\n", + "5.000000 1596\n", + "3.373038 528\n", + "Name: count, dtype: int64\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "print(eda_df.bedrooms.value_counts())\n", + "ax = sns.barplot(x=eda_df.bedrooms, y=eda_df.price)\n", + "ax.set(\n", + " xlabel = \"Bedrooms\",\n", + " ylabel = \"Price (in Thousands)\",\n", + " title = \"Number of Bedrooms vs Price\"\n", + ");" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Observations\n", + "\n", + "As the number of bedrooms increases, so does the average price, but the most popular choice among customers was 3 bedrooms." + ] + }, + { + "cell_type": "code", + "execution_count": 302, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "price_vs_column(\"sqft_living15\", \"Average sq. ft of Nearest 15 houses\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Observations\n", + "\n", + "As the sqaure footage of the surrounding houses increases, so does the price of the house." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Modeling" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "First, we make a copy of the existing dataframe." + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "metadata": {}, + "outputs": [], + "source": [ + "modeling_df = eda_df.copy()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Location columns are dropped so as to preserve the privacy of the home owners." + ] + }, + { + "cell_type": "code", + "execution_count": 304, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
pricebedroomsbathroomssqft_livingsqft_lotfloorsviewconditiongradesqft_abovesqft_basementyr_builtyr_renovatedsqft_living15sqft_lot15
0221.93.01.001180.05650.01.00071180.00.01955.00.0000001340.05650.0
1538.03.02.252570.07242.02.00072170.0400.01951.083.7422151690.07639.0
2180.02.01.00770.010000.01.0006770.00.01933.083.7422152720.08062.0
3604.04.03.001960.05000.01.00271050.0910.01965.00.0000001360.05000.0
4510.03.02.001680.08080.01.00081680.00.01987.00.0000001800.07503.0
\n", + "
" + ], + "text/plain": [ + " price bedrooms bathrooms sqft_living sqft_lot floors view condition \\\n", + "0 221.9 3.0 1.00 1180.0 5650.0 1.0 0 0 \n", + "1 538.0 3.0 2.25 2570.0 7242.0 2.0 0 0 \n", + "2 180.0 2.0 1.00 770.0 10000.0 1.0 0 0 \n", + "3 604.0 4.0 3.00 1960.0 5000.0 1.0 0 2 \n", + "4 510.0 3.0 2.00 1680.0 8080.0 1.0 0 0 \n", + "\n", + " grade sqft_above sqft_basement yr_built yr_renovated sqft_living15 \\\n", + "0 7 1180.0 0.0 1955.0 0.000000 1340.0 \n", + "1 7 2170.0 400.0 1951.0 83.742215 1690.0 \n", + "2 6 770.0 0.0 1933.0 83.742215 2720.0 \n", + "3 7 1050.0 910.0 1965.0 0.000000 1360.0 \n", + "4 8 1680.0 0.0 1987.0 0.000000 1800.0 \n", + "\n", + " sqft_lot15 \n", + "0 5650.0 \n", + "1 7639.0 \n", + "2 8062.0 \n", + "3 5000.0 \n", + "4 7503.0 " + ] + }, + "execution_count": 304, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "modeling_df.drop(\"lat\", axis=1, inplace=True)\n", + "modeling_df.drop(\"long\", axis=1, inplace=True)\n", + "modeling_df.drop(\"zipcode\", axis=1, inplace=True)\n", + "modeling_df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "A function that shows correlations of a particular column." + ] + }, + { + "cell_type": "code", + "execution_count": 305, + "metadata": {}, + "outputs": [], + "source": [ + "def column_corr(column):\n", + " return modeling_df.corr()[column].sort_values(ascending=False)" + ] + }, + { + "cell_type": "code", + "execution_count": 306, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "price 1.000000\n", + "sqft_living 0.581669\n", + "grade 0.573017\n", + "sqft_living15 0.508827\n", + "sqft_above 0.474044\n", + "bathrooms 0.424381\n", + "bedrooms 0.290729\n", + "floors 0.274164\n", + "sqft_basement 0.222340\n", + "view 0.203565\n", + "sqft_lot 0.086452\n", + "sqft_lot15 0.067868\n", + "yr_built 0.061169\n", + "yr_renovated 0.030432\n", + "condition -0.013279\n", + "Name: price, dtype: float64" + ] + }, + "execution_count": 306, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "column_corr(\"price\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "This function creates a histogram for the given column." + ] + }, + { + "cell_type": "code", + "execution_count": 59, + "metadata": {}, + "outputs": [], + "source": [ + "def column_hist(column, axis):\n", + " sns.histplot(data=modeling_df, x=column, kde= True, color=\"purple\", ax=axis, bins=10)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We then use it to view the distribution of columns of interest." + ] + }, + { + "cell_type": "code", + "execution_count": 60, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import warnings\n", + "warnings.filterwarnings('ignore')\n", + "\n", + "fig,axs = plt.subplots(2,3, figsize= (17, 10))\n", + "column_hist(\"price\", axs[0,0])\n", + "column_hist(\"bedrooms\", axs[0,1])\n", + "column_hist(\"floors\", axs[0,2])\n", + "column_hist(\"bathrooms\", axs[1,0])\n", + "column_hist(\"sqft_living\", axs[1,1])\n", + "column_hist(\"sqft_living15\", axs[1,2]);" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "None of the columns display a normal distribtion and are skewed. This can affect the performance of the model negatively." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Simple Linear Regression" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "As living area square footage has the most correlation with price, we ues it to build a simple model." + ] + }, + { + "cell_type": "code", + "execution_count": 307, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Model is significant: True\n", + " OLS Regression Results \n", + "==============================================================================\n", + "Dep. Variable: price R-squared: 0.338\n", + "Model: OLS Adj. R-squared: 0.338\n", + "Method: Least Squares F-statistic: 1.101e+04\n", + "Date: Thu, 02 May 2024 Prob (F-statistic): 0.00\n", + "Time: 02:42:59 Log-Likelihood: -1.4049e+05\n", + "No. Observations: 21534 AIC: 2.810e+05\n", + "Df Residuals: 21532 BIC: 2.810e+05\n", + "Df Model: 1 \n", + "Covariance Type: nonrobust \n", + "===============================================================================\n", + " coef std err t P>|t| [0.025 0.975]\n", + "-------------------------------------------------------------------------------\n", + "Intercept 169.1084 3.169 53.356 0.000 162.896 175.321\n", + "sqft_living 0.1554 0.001 104.930 0.000 0.152 0.158\n", + "==============================================================================\n", + "Omnibus: 1394.672 Durbin-Watson: 1.963\n", + "Prob(Omnibus): 0.000 Jarque-Bera (JB): 1681.789\n", + "Skew: 0.660 Prob(JB): 0.00\n", + "Kurtosis: 3.360 Cond. No. 6.04e+03\n", + "==============================================================================\n", + "\n", + "Notes:\n", + "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n", + "[2] The condition number is large, 6.04e+03. This might indicate that there are\n", + "strong multicollinearity or other numerical problems.\n" + ] + } + ], + "source": [ + "from statsmodels.formula.api import ols\n", + "import statsmodels.api as sm\n", + "\n", + "formula = \"price ~ sqft_living\"\n", + "model = ols(formula, modeling_df).fit()\n", + "summary = model.summary()\n", + "print(\"Model is significant: {}\".format(model.f_pvalue < 0.05))\n", + "print(summary)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Observations\n", + "- The model only describes about 33% of the variation of the price and thus is not efective.\n", + "- An intercept coefficient of 169.1084 (in thousands) tells us that a house with 0 square foot living area will cost $169,108.40.\n", + "- With a coefficient of 0.1554 (in thousands), we can expect the price of a house to rise by about $155.40 per square foot added.\n", + "- Both the intercept and sqft_living coeffiients are statistically significant as shown by their low p value scores.\n", + "- A Kurtosis score of greater than 3 indicates that the dstribution has heavier tails probably due to outliers.\n", + "- A skewness of 0.660 tells us that the distribution has a moderate positive skew. This means that the distribution has more values extending towards the higher end." + ] + }, + { + "cell_type": "code", + "execution_count": 72, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Mean absolute error: 132.19247328012102\n", + "Mean absolute percentage error: 0.27535308256314295\n" + ] + } + ], + "source": [ + "mean_absolute_error = model.resid.abs().sum() / len(modeling_df)\n", + "mape = mean_absolute_error / modeling_df.price.mean()\n", + "print(\"Mean absolute error: {}\\nMean absolute percentage error: {}\".format(mean_absolute_error, mape))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "With a mean absolute percentage error of ~27.5%, the model is right more often than it is wrong." + ] + }, + { + "cell_type": "code", + "execution_count": 73, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Plot actual results and predicted\n", + "sm.graphics.plot_fit(model, \"sqft_living\",)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 75, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAj4AAAGtCAYAAAAWKH7cAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAACNGklEQVR4nO3deXhTZdo/8G/WJt3S0rQF2rJ1YRGQ0kIBl0JlU6Gig844yMy4o7yi6LggrvgqOirM6KujMML8BhlFGZFFFIRCRwUKlFVZurC1BdumNN3btMn5/VFOyJ5zkpPkNLk/1+V1SXOSPKdJc+48z/3ct4RhGAaEEEIIISFAGugBEEIIIYT4CwU+hBBCCAkZFPgQQgghJGRQ4EMIIYSQkEGBDyGEEEJCBgU+hBBCCAkZFPgQQgghJGRQ4EMIIYSQkEGBDyGEEEJCBgU+hBBCCAkZ8kAPQKzq6poglmYeEgkQFxclqjH5Sqica6icJxA650rnGXxC5VyD5TzZ83CHAh8nGAaiewOIcUy+EirnGirnCYTOudJ5Bp9QOddQOU9a6iKEEEJIyKDAhxBCCCEhgwIfQgghhIQMCnwIIYQQEjIo8CGEEEJIyKDAhxBCCCEhgwIfQgghhIQMCnwIIYQQEjIo8CGEEEJIyKDKzYSIiNHE4EhVA3TNBmgjlRiVpIFMKgn0sAghJGhQ4EOISBSU6vBuQRlqmg3mnyVEKvFUXhry0rUBHBkhhAQPWuoiRAQKSnV4dtMJq6AHAGqaDXh20wkUlOoCNDJCCAkuFPgQEmBGE4N3C8pcHrNsVzmMphDoHkgIIT5GgQ8hbhhNDIor9Nh2sgbFFXrBA5AjVQ12Mz22qps6cKSqQdDnJYSQUEQ5PoS44I+8G52boIfvcYQQQpyjGR9CnPBX3o02UinocYQQQpyjwIcQB/yZd1Pf1un2mMSoMIxK0nj9XIQQEuoo8CHEAX/l3RhNDJbvKnd73BO5g6ieDyGECIACH0Ic8FfezeFK9wEWAMSGK7x6HkIIId0o8CHEAX/l3ehaKLGZEEL8iQIfQhwYlaRBgpugRoi8G20EJTYTQog/UeBDiAMyqQRP5aW5PObJSalO82641v7JTPZPgEUIIaQb1fEhxIm8dC3eyh9mV8cnMSoMT05KdVrHh0/tHzbAenbTCafjuG1Eb+w4XUtNSwkhRAAU+BDiQl66FrmpcZw7prO1f2yxtX/eyh9mF/w4C7A0qu4/zxV7zpt/Rk1LCSHEOxT4EOKGTCpBVkqM2+O41v7JTY2DXGYdONkGWBX6NnxsEfCwXAVQhBBC3KMcH0IE4m3tHzbAmjw4HhuOXXL5ONS0lBBCPEOBDyECEar2DzUtJYQQ36GlLkIEIlTtn1BrWtrc3oWXvzuFqoZ2JGlUeHX6EESq6KOJEOIb9OlCiEDY2j+uZmu4bE0Ppaalf/z0EE5UN5v/Xa5rxaQP9mBYYiT+3z2jAzgyQkiwoqUuQgTibe0flr+KJwaabdBj6UR1M/746SE/j4gQEgoo8CFEQOzWdNvAJTEqjPNOLKECKDFrbu9yGvSwTlQ3o7m9y08jIoSEClrqIkRgfGv/OHsMT4on9hQvf3eK83Hvzhru49EQQkIJBT6E+IBl7R+jibELgmzr+DgiRAAlRkYTg7LaFk7HVjW0+3g0hJBQQ4EPIT7krH3Fn/PScJc2yu39uRZP7Ckc/T5cSYoO8/GICCGhhgIfQnzEVfuKZzadQHS0Ctm9IwMwMs84mrniM/tUUOL49+HKHdf24TtMQghxiQIfQnyAS/uKVzefwNf3j4FUIv6lKz6NVwH7ICkvJgLvuPl9ONLcYfJq3IQQYosCH0JseDuzAXCrvnypoR2HKxtEv5TFt/GqoyCp1zencLmFf8HFMIVnYyaEEGdEuZ1dr9fjmWeeQU5ODsaMGYNHH30UNTU1AICjR4/izjvvRGZmJvLy8vDll19a3XfDhg2YMmUKRo0ahTvuuAOHDx8OxCmQHqqgVIf8lUWY98UxvLD1FOZ9cQz5K4tQUKrj9ThcqyrvLNGhuEIv2r5bXBuvsuNngyTboM+ToAcA/v7jOY/uRwghzogy8HnsscfQ2tqK77//Hrt27YJMJsOLL76IhoYGPPTQQ5g1axYOHDiA119/HUuXLsWxY8cAAEVFRXjttdfw5ptv4sCBA8jPz8cjjzyCtra2AJ8R6QmcXbTZmQ0+wQ/XqspfHrmIeV8cw8wV+7By73lsO1kjmkDIaGKw7nAV575hXIIkvnQtnYI+HiGEiG6p6+eff8bRo0exZ88eREZ2J36+9tprqK2txfbt2xETE4M5c+YAAMaPH4+ZM2di7dq1GDlyJL788kvceuutyMrKAgD86U9/wrp167B161b85je/Cdg5EfHjOrORmxrHadmLS/sKS7UtnVix57z5367yZ2wJsTRni+/uK12zgdPyHl/aCMcBpNHEoLhCj0OVejAMkJ0Sg9EpMT1+qz8hxPdEF/gcO3YMaWlp+OKLL/DZZ5+hra0NN9xwA5599lmUlpYiIyPD6vi0tDSsX78eAFBWVmYX4KSlpeHUKW7F0iyJKd+UHYuYxuQrgTpXPh3Rs/vFuH08uUyCP+el4Rmeu5hY7CzTwxP6475x/ewu6EYTg8OVDSgsq8O3J2ugb7s6M8Jul8/L8KzIoSe7r7SRSug8XM5y5b3bR9i9FwpKdHh9ewkaLKo6ryqqgFImwR/HpuCB8f0DEgCFyt9pqJwnEDrnGiznyXX8ogt8GhoacPr0aQwfPhwbNmxAe3s7nnnmGTz77LPQarVQq9VWx6tUKrS2tgIAWlpaXN7OR1yc+xor/ibGMfmKv8+1o7KR23FSKbQc6u8AwF3aKERHq/Dq5hO45GEhvo/3nMfGn6vxSv4wTB/evbX7u58vuXzM2itB09/vGW2+D1dGE4Plhfs5Hy8B0FujwpRRydh/9jKn+/SKUOAyxyWsoktNuD8t3vzv736+5DSYNBgZrNx7AWuLq7DsrmsdnrvRxGD/2cuoaWpHQpQKYwf2EjxICpW/01A5TyB0zjVUzlN0gY9S2T21vXjxYoSFhSEyMhJPPPEE7rrrLtxxxx1ob7f+sG9vb0dERAQAQK1WO7w9NjaW9zjq6prABD7NAkB3FBsXFyWqMfmKr8+VnSnRtRigjVAiM7l7WSjMxG3bdJjJBJ2uifPzZfeOxNf3j8HhygbsLNHhyyMXeY/518Z2zPv0EP6SPwwA3M4isb+2lzf+gsyECF4X9oMX9LyCNAbAwtxBqL/cjEFRCk7d6TfcPwaLNp9EYXmd28ff8cuvuG1od+BjNDF46euf3d6n1WA0/74sZ70KSnR4x0kxSU9nxyyFyt9pqJwnEDrnGiznyZ6HO6ILfNLS0mAymdDZ2YmwsO6qraYrF6WhQ4fi3//+t9XxZWVlSE9PBwCkp6ejtLTU7vYbb7yR9zgYBqJ7A4hxTL7ii3MtKNXhnZ2lqLWYbYiPUODPN6UjNzWO00V7VJKG97ikkqvVlz0JfFj/u70EEh5PXt3UwXu7PNfdaAAQq1bgmZvSMCldC4bpPs+n8tIcLpNJ0B0kPTkpFQqZFMP7RHIKfPaeq8fOEh3y0rU4XMkvh+idgjLceCUny10xSa4NZLkIlb/TUDlPIHTONVTOU3S7uiZMmICUlBQ8//zzaGlpweXLl7F8+XJMnjwZM2bMgE6nwz//+U90dnZi37592Lx5szmvZ/bs2di8eTP27duHzs5O/POf/0RdXR2mTJkS4LMigcZe+GptllhqWzrx7KYTKCyv83lHdDbh2VON7V1o6DDyug+fQAbgvhsNAOrbOrF8d7nVbjdn3el7a1TdMzBXgotLjdxnldjt8nzPpeZKwjXfLfmEkOAmusBHoVBgzZo1kMlkmDZtGqZNm4bevXvjjTfeQGxsLFatWoXvvvsOOTk5eOGFF/DCCy9g3LhxALp3eb388st45ZVXMHbsWHzzzTdYuXIlYmJiAntSJKCMJgavby9xecwb20uQmxrn8KKdGBUmyIyATCpxG1wJjU8gA/APzhxt9c9L12LTgzn46K6R+N9bhuCju0bix2fzrJaTqho6OD9HdVMH1h2uQq9w/tUMue42YxPXCSHBT3RLXQCQmJiI5cuXO7xtxIgR+Pzzz53e97bbbsNtt93mq6GRHuhQhR6NFjuAHGlo78KhCr3PO6KzMyJ8top7il2a44MNzvju6rLd6m/ZXFUigd3vL0nDr/no8t1nkBCpRLRK7va1tKSNVHKeKeI7o0QI6ZlEGfgQIqSDFXrOx43pH2t10fZFjZy8dC0mpsXhTFMnyi/qUVHfhq+OXrRbhvOWp0tzngRn7IwJ13yivtEq3uPiGyjGqOQY0Scaxy9x27HHd3aMENIzUeBDgh7X2g529WJ4NubkQyaVYHxqHNI1SjAMcG9OP6wuuoCPLYoYeioxKgxPTkr1aoyWM19cd6PxmTGpbvY8yNOo5DB0mdDW5Xonnr69C7d/sh8LJ6ZyTlwHfBPsEkLEgwIfEjScXbBGJ8fgE1S4vf/o5Bjz//NtzOktmVSCB8b3xyBthEfLYJFKGUb0jca4AbGYfW1fKOX80/cc/f747EbjM2MSFy7jPT5WQ3sXPpg9AkeqGvCvA5XocBEA1TQbsGjLSczNTsaag5VOj5s1ojcA4YJdo4nBoQo9DlboIZF0v7eyqLI0IaJAgQ8JCq4uWLmpcdCo5FaVfm1pVHKr5S0h21fwwc60FFfosWjzCTS62MUVq1Zg6pB4bDtVC31bJ/aeq8fec/VYe7DS7YXa0GXC+qMXUalvR3KMCtpIJf62+4zT3x+fGRMufjpbz/lYR+pbO/HQhAG4f1x/HLhQj+c2n0SLwfnvavvpWiydORTLd5U7PI+P95zHmoOVaHXwGGywe/foJOSmxbmdASoo0eF/t5dY5SJ9ggpoVHI8PzVD0ICZEMKf6HZ1EcKXu+aiheV1eH5qhpN7d3t+aob5YhboXUAyqQRj+8di8bTBLo+bcU0i1h2+aNWuArh63st2lTtsePpe4Rnc8N6PWL77DL48chHLd5/B4i2nXP7+hN7qX+1lIjE7uySTSqCQSV0GPUD36xWrVmDTgzl4aEJ/h8c4CnosfXaoCvO+OIb8lUVOG9aylaUdJWA3tHfxbnZLCBEeBT6kR+MzO+Nwq3qk/VZ1sewCclYTJzEqDEtnDMW2UzUu7+/oQv1e4RmsOVgJPiVrXP7+PNzq30vt+VKX7ewS39fr62OXPH5uwPEWfqD7vfjKpl/c3n9ZAdUMIiSQaKmL9GhcZ2fWHa5CXLgSr0wfDAbdSyXOEle55qr4YxeQs+31fDqhsxfq12cMwdpi53kuzrCzW0Ju9b+k96x3GWA/u8T1dThT14J1hw2ClRGwXe48XNmAXxvd1yeqbua3A44QIiwKfEiPxvXb/vLdZ8z/z+auOLvwsEX8hMxp8Ybl9npWYZn7dg+2lm4v5TXTY4n9PbNjYROhd5yu5R0AGU0Mmjv5D8TZbjUurxfQ3cFdSLZb+Pl0p6eaQYQEDgU+pEex3XnkSTVfdgZk6cyhiFUr7GYvuBTx87Z9hTeMJgbfnnS9zOVIs5scFld6hStQXKGHrtmAC/o2bLCpO8R15xObhG7kEffcl5OCsf1j7YIry/fC7SP7CFIKgC/LAEYbwX0GkGoGERI4FPiQHsPZzi2+1XxZi7ectJoBsbx4OyviJ0SNHG8dqWqwS2j2JY1Kjle+O+1yRoXLNv+CEsclAtx5aMIAuyDT0XshXCFFa6fr2j5Cswxg9G2d5masriRG+m+2kBBijwIf0iO4qqvjKdtlH9uLN9+cFn8VvvNmmYTLhdmWqzIAtpxt8zeaGLzjJgndEYXMvt2Fs/eCv4Mey+XOglIdntt8ktP9RiZF+3JYhBA3KPAhosdl55ZGJYdSJhGk7YPlxZtr+wpfVnm25c0yyYSBsZxr6EgARLupf2TLWeuK/WcvexSkdhq7f7fs75DLe8Ff2OVOvmP6/nQtfjp7GS9OzcDkwfE+HCEhxBEKfIjocdnBxFbzlUklqGnqwLGLjVh/1LNty7YXb6OJweqiC/jsUJXVkhob2ADwa5Vnrsm8jswdk4JBcRH49GCl25kfBvxme1iOZqRqmjzfxWX5O+Szm80b7JImALfLnYcq9LzH1GowYtGWkzjxaxMW5A4SbuCEELco8CGix3Vpp761Ewq5FP/3w1mvL47scxaU6vC6TRVeFhvYaFSu/4yErvLsaQf1WLUCI/pEIyslBqnacLzyXYkg47HlaEYqIYp/U1JL7O/Q17uhbh4aj74alVWLCVfLnez7w1NrDlZiWJ8oTErTUn8wQvyEAh8ielyXdi7o27BCoJ09F/RtTnNJbLmbFeHbuZwLTzqo17d14vZP9uOpvDT09qA7OheOtvkbTQxMJsbjJHTg6u/Ql7uhpBLg25O1ALpbTFguVTp67bi+P9z53+9OY3mYdSuNGLUCNw9N4NQigxDCDwU+RPQ41dWJDMOGo+4baQLcEny/PnoJDO80YOd8MVPhKPm6vq0TywrKnOY6WW7l93S5zBXbbf6Ocp88pWs2YPLgeJ+MG3Cf7G5JyFyjlk4TWjqtz0ff1onPDlXhs0NVDnPFqIM8IZ6jlhVE9NilHVdmjezNKbE5QinjFM7UtBgESZRm+Wqmgk2+njY0AVkpMThxqQl1re7H/dfdZ7BwYqpg43DUusJZDzVPaSOVnN4LQlu2y77FhL9yjQD7FhkFpTrkryzCvC+O4YWtp9z2DzOaGBRX6LHtZI3D3m2EhBqa8SE9gru6Op1d3LYyu2tm6Qv+qvLM9uHiorqpA7HhCiydMRRv7SzjXRcoMTIMs0b2RkqM2uGMg9C7ryx/h+x7wVnuldAcLVUGovLysl3lMJkYLNpiv23eWQd5f+42JKSnoMCH9Biu6uoUV+gDNi6Nmy3fQlR5dre00WYw4lOOQQ+rsKwOO0tqeQc9CycOwm8zk1yek9AzIk/kDrI6/9zUOETNGIpH1x8X7DlcsQ10AlF5ubqpA2/tdB1MWi6PTRuS4DAQ9tVuQ0J6Cgp8SI/iqG8V4N0Wb289PzUDgPttz1w4CnAKy+vsHjtaJcfdo5Nwb04/FJbX4bVtp3lnJH12qIrnPbr92tjhNpATakZEKZMiNy0Oy3eX281aLJyYyuk1T4hQYta1fVDf2okvj3DLA7NlG+gE6v3GNUitaTa4nf0TerchIT0FBT4kKHi6xdsbGpUcz0/NQF66FkYTg0ilDIcq9WAYIDslBqOvbIdm2QY1U3pFWj2eo2UJZ7NJje1d+HjPefzrQAXaPKhY7EkFZ9Z3J2vweO4glxdMoWZEDEYTvj9da/fzmmYDFm05ibnZyS4v8A9N6I/7cvqZZwU9CXwcLVVyeb95s4vNH3yx25CQnoACHxI0nOUBxaoVqOf4TZnrxWpY7yjMv34AslJiHAYsW36ptsqjcHRMr29O4Zm8VNyUEe90a7S7rfKeBD2A50EP0L0t3t0Fc0SfaEgl9julhLb9dC2WzhyK5bvK3c62eTpL42ypMi9di6Uzh+Kt70uht3idLIsfChmI83kfc0Vd4kkoosCH9Ahct+86ygMa0Scat3+y3+UFT6OSY+mMoWAAzOeQN3Li1ybMX3/caaBkmUcBOL4AXm4x4LnNJzEnq9HhrIaYubtgHr/U6POgB7iSpK1WYNODOd15RU0dqG/rRKxaAY1KDqOJMb9P+M4KuksCLijVYfmucqugp1eEEgsnDjLf5+7RSR4vKdp6ZnKaXYDnLeoST0IRBT5E9PjuTHGUB+Tugvf81AyM6R8Lo4nhNSvgbnZoWUE5TIzrWZm1xcJcGLnyZpmL5e6C6c+ZBF2zATKpBA3tXXZVu23fJ3x2hDGM89+Ssxm6+ivB7N2jG5GbFofrB/XyOvCxPAepRCLYLJK/dhsSIjZUx4eImrNaMLa1TdxhL3gJNhds2/ozQteJqW7uELQekBC8DXq4XDD9OZOgjVTyep/kpWux/ZHxeHhCf0S7aDdS29Lp8D3maqs++7v97FAV5n1xDK98e8ptSxN3Xp4+2Cpwu3t0klePxxJityEhPRHN+BDR4lIL5t2CMqudKa6WxFxth7fkSTuIUMLlgjkqSeN2m78QEqPCMKJPNGb9o8jlcbY7mGRSCQZpI6Dk8NVv2a5yXD+wF45faoSu2YC6VgPn94UQQW+9TUHK3LQ4QZbPynUtaDN04edLTQCAlFg1Zl/bF0q5818K+/fVUdmIMJOJKkaTHokCHyJaXGrB1DQbsLroAh4Y35/Tkhi7DMZ+gO84XeswALIMknaW6DzeBt1TPTShP74+dsmr7fn+qA+cro3A0h2lbgMM2x1MfPpsVTd14NYVRbzrHVnSqOQIk0s9Srr31VZ6R33t/rr7DCYPjsdrtwyxC2ioGCIJFhT4ENHimify8Z7zaDUYORdr4/oBbpkr5Gng44udOP7QL0ZtThi2TBI/fqkR207WuO0PdaSqwS9buX88e5nzsez7yZOq0t4EPUD37rwPZ4+AVCox/z6HJUYhf2WRVXK0LWdb6RdOSsWizfYVnL3FAPj+dC0Ky3S4N6cf7r1SCsBZoEjFEElPRIEPES0+eSJri7kVayssr+P9Ae7NN+zpQxME29UzLDESlfo2NHb4vu1Ghb4NMqkEo5I0OFLVgMKyOjy3+aRVAODq274Yt0nvLK1FXasB+tbOgCxhXm7txLShCQC6g+/Zqw+4DHoAx8uK7G4yXzIYGXy85zw+O1SFRZPTsXRHqcvjqRgi6Uko8CGixSfgcLd1urqpA4cq9G6/6Tv6AOeyDdo2n4VdFtKo5IIFPtOGJuC3mUnmWZgL+jZsOHrRapnHsoaMN/WMvj72KwbEhbvcPu0qWIwNV/A9PZ/bVVqHXaV1AXt+NpDnsswWrpBi7pgU5KbGWeWtVejb8LGDJSpfaWzvctgbzBYVQyQ9CQU+RLSErsZcdL7ebRDl7APcWcKzRiXH70Yn4Q9jUszJr5bLQHy3xzsjlQCzr+1rt1X/vpx+TpO1PalnZP49NHdwXkpxFCzS935r7JIV12W21k6TecZFAveFLMVAjLN8hDhC29mJ6IUrhHmb/pvjzIuzD/C8dC02PZiDhyb0R3SYDED3BenjPecx6x9FOFTZYHcfmVSCaUMSPB/0FXOykh3utmEDoWlDE5Bl0yLD9jalXCroVn0WGyxautza8/KabMVw3Ia+cOIgt1vM2SUrvs1bG9u7ekTQA1yd0TKaGBRX6LHtZA2KK/Qw+qOSJSE80IwPES0+O2+46DRy+wB2lVtUWF7ncDdMbUun1c/Z/BcAbptFuiIBcE92MhbkDuJ8H6OJwaEKPQ5W6CGRAKOTY8xBUV66Fg9N6O/wHLwhhu7lQlPIJAhXyNDa6TynKjEqzNypPjNZg+WFZ3Cpod3qdsudcME6KyJB97Z7RxsH1AopbsqIx6LJ6S63yhPiLxT4EFHyZOeNUJzlwfAZE5v/ws4Meerl6Rm49ZrenI8vKNXZVSX+BBVWDVXvy+lnt1XdW2LpXi4kLjV4pg6Ov1onKkOL34wbgO+PVDqtExUMAaEjDOA0F6it04Qtv1Tjm1+qHQbxXNvRECIUCnyIKB2u5LckIKS/7j6DSWlauw9fvssUALzegdU7WsX5WFczZA3tXVaJyELmTmlUco+6lweDNQcroVbK0C9GDW2kElN6RSK7XwzYbhfsso9lnlVPDwg9xeDq7Ccb/FBtIBIINO9IREnXErgLg6OcFcD/yxSOAgpnjCYG7+x0veUY6O4dZjQxTlt4eKKhvQuF5XXmcbD5HRqVHEtnDnXYJiRWzf87V5hMnLMAK/acxwtbT2HeF8dw/VsFKCjpbnFRUKpD/soizPvimPn22z/ZL0jOl1BuGBjj9+dcW1wJQ5dJsHY0hPBFMz5EVIwmBnvL63CmroX3fR1VG/ZUTVOH1Tf1UUka0S5TGE0M1h2u4rQ0U918ddeabQuPM3UtWFVU4dEYlu0qh8nEYPnucrtv709OSkX/3hqUX9RDG9H9u7znXwdQ38YvabeDY45WIP3a0I5nNp3A3OxkpwU11xysxA0De+EHHsUX+chK0aC4wj5wt6VRyXHsUjOnx3x4Qn/8u7gSTQLUkDIxwBdHqvCZm+a8VBuI+AoFPkQ0HE17c5UYFYb7cvrhvpx+WHe4Cst3n/FqLMt2lVsVl0uIVGLhxFS/LlM0tHe5rY3iye/McubKcnt8cYXe48CnuqnDYY5HTXN3t/KP7hmN6UMTzEtAYXLvcp/Eig3N3BXU9FXQA4BT0ANw3yKfEKnEgF7hggYgRyobPS4tQYi3aKmLiIKzaW+u2O3CMqkEv81M8noJx7aibk2zAYu2nOS9TKH2ciu+q+U1T39nzmau2IRkX3h18wmrbc18Z3t6GjHv4E6MDHPZld7W5dbu975ewNcsXMkt8A3WXXAksCjwIQHnzQ6uxKgwu8rBbGKtL2w/Xeswb8WZuWNSvHo+Z0GKp7+zxEj73k8sX/7eLjW047BFnSNvd7ux/jAmGWP7xQjyWMEsVq3AkpsH46O7RuLl6Rm8+qh1mYQdi1QC3DKM2xcIsS4vk56NlrpIwHHdLXX/uBSMTo6BBN0F8lxtfXVWaTlWrcD0oQmIVsmxwSYfiEtLh+qmDsSqFVYNPCv0bXaPxdZvyU2N8yrvqK7FYJdr5EkhPNaTefa9nyw5+71JJd7PYlgmrOemx+FUbat3DwjgXwc8r5EUSurbOpEQFYaslBhsO1nj0+cKk0nRYXQeLc3JSsaYfrFul41dBemEeIMCHxJwXKezB/aKwNj+sZwf1zZ51zZQutem3UNNUwde+vY07/FmJmuctqwA4NW27he3nrIKOKLDZPhdVjKSNdy3uQNAhFKGl6YP5rRF2NHvrb6t0+tu4HUtBhhNDGRSCc7WtXn1WIS/nVd2m52r9z7gdCYxKgxf3TcGr3x7CjtKdLCMlaWS7qCH3cru7u+ircuIwvI62tZOBCdhGEbEq9GBo9M1QSy/GYkE0GqjRDUmIRVX6DHvi2Nuj/vorpE+TXTkOo6HJ/S3m+GJj1Dg9mv7muu5sN9U2eChsEyH70uE254boZShxcB9h82Smwfj5mGJXj2nq0RqrjNCbI2Wj346S8FPELJcdjZ0mbD+6EVU6tuRHKPC7Gv72lVu3lFSi//97jRaOp3PEDlqggv4p/BhsH/2soLlPNnzcIdmfEjAjUrSID5C4XI7NtvkkQtPPxC5VBvWqOQOu2PbtqzQqORgAF65FHzwCXoAICEqzOHP+fyu8tK1MJkYh7u3uC6DsTVaIgTqv0bE4+EJ/a0CFKVcit9nJTs9vqBUh+W7yl0GPYDjbe2OgvAYtQI3D01AblocVX8mLlHgQwKusLwOBjc1WthdW+44+kCMVslx9+gk3JvTz+VjCFltWEyNJZ0FjY5+VxFKGUb2jca4AbF239CNV+r0CMHdxc7XhiVGQtdiCMkKyr6QEKnEvTn9OB/Ppw+f7bZ2Z/fVt3Xis0NV+OxQFWJUctw8LJGCIOIQfe0iAcV+iDkLFDQqudOpbmePZXsxa7zSQX3q3/c6rAZrWW34jK4FGgdbfTUqOR6a0F9UAQ3LXfd6R0Gjs99Vi8GIvefqsXz3Gdzw3o94r/BqPSRPE6rF6ER1M269JgG3DI3HsN5RUNB10StP5aVxDi482ZHI5tVxva++vQufHarCvC+OIX9lEVWBJlZoxocEDJcPMZVchtzUOEEeq9GmXxXAvQBgQ3sXmkQY9ADAzOG9EaNW4LNDVVZLa7adwVlGE4PXt5e4fVwTY91bKdhqqqwuoh1hQpAAuH5gL87HexJAs9vaPbkvu7zK9QsUCX4U+JCA4fIhZtliwdvHYrE5A4XldbyWtTYdv8T5WH/afqoW384bZ7dLzdkU/+qiC7xyj9YWV2LedQOopgpxiAGw/uhFl/k8lgrL6ng9vuVSrTfBN7XAICxa6iJ+ZbmsdOBCPaf7uPqwYx9vJ48dU9VNHThUoec93R7ovBRn6ts6ceBCvV3QA8D8uy6u0MNoYmA0MfjskOseSbZMTPeFzZeVnbniU3GY+M+es5fN7zFXjCYG356o5vXYeelaHKlqgNHEeBV8O2s+TEIPfYoQv/G0F5ezDztvenutO1wlinyVGJXcrj2GJxb852ermimOdpUlRCoxa2Qfj3aaVerbIZNKsHBSqtf1fDyhUcnx2i2D8fQm/z83ca/ovB5F5/XmcgXOlpSOVDVwfr+zJRLYhGUh+uUF23It8QwFPsQv+OzisORqR5I3u68Ky33XJNKdB8f3Q/9e4ajrMOHTvWcFeUzb79mOkrBrmg1WW+55PT7DYMfpWsF2dfGVP7w3Xtx6Gh1C908ggmLzaZbOGIrYcIXdsiufwMN28ojtlzcnKwlr3XR2d6ZXuMJhJXQSWijwIT7nTS8uRzuSvHk8MchM0qDZYMRfd5yyC1jEav3RS1h/1P85TpowGfJH9DEnWYeamCszd1x2E4YrZGjt5FffyVcWf3PSKnBhZ4KEyBPju1TL0qjkeOW701azRe5mqEhwEnWOj9FoxNy5c/Hcc8+Zf3b06FHceeedyMzMRF5eHr788kur+2zYsAFTpkzBqFGjcMcdd+Dw4cP+Hjax4ek2aGf5HD19W/WBC3r8ZUdJjwl6AunG9F7Ydsq3vaXE7PaRfTiXUIgMk+HBcf0QpQz8x7qj2ZpnN51AfWun13linvaMa2jvsvvcYMdF291DS+D/Qlz4v//7Pxw8eND874aGBjz00EOYNWsWDhw4gNdffx1Lly7FsWPdbQaKiorw2muv4c0338SBAweQn5+PRx55BG1tVBo/kDxdV2e3n9t+KPX0dfrdZTroWrnn2SRGhWFudnKPTeyNVStw64g+iPKgI3th6eUeHeR6i+ERHtc0G3CpqQMSqfXHupALOSq5FOnaCI/v/9fCM1g4KVXAEXHjrlL4sl3lThOzjSYGBy9YbxIgPZtoA5+9e/di+/btmDp1qvln27dvR0xMDObMmQO5XI7x48dj5syZWLt2LQDgyy+/xK233oqsrCwoFAr86U9/QmxsLLZu3Rqo0yBwnpzM1Zvfl8JgkdvR07dVn73MLxB/YuIgLMgdhDdnDPXRiIR167AExFgEafVtnfjm+CU0dfBfhmn04D5ixiedJFwhxdfH+e2A2vJLtV3yureX6YG91MhNjcPjuQOxc/4EPJXneeBS3dSBWLUCb+UPs5v5SYwKw0MT+ns5WnuxaoXbHZnOdnwVlOhw/VsFmPfFMbyw9RQVRAwSogx86urqsHjxYrz77rtQq9Xmn5eWliIjI8Pq2LS0NJw6dQoAUFZW5vJ2PiQScf0nxjFx/S8z2btt0PVtnbh1RRF2lerMjxfTQ2c/PLF8VzlMDIOsfjGCbCf3ZOaFj29O1AiyUy0Y8Sn019ppgr7Nef86fzl7uQ2F5XX4W+FZ3P7JfjS0d6KPRuXx4+laDLgpQ4vND+Xgo7tGYsktg7Fw4iA8ct0AqBVSwWc2pw9N4Dwuy8+tXaU6PLPpBC41tFsdxy6PsZ9HwfIfEPgxCHUe7oju6mEymfD000/j3nvvxZAhQ6xua2lpsQqEAEClUqG1tZXT7XzExbnv8OpvYhwTV7ePTsLH//V8B5O+rRPPbjqBv98zGtOH98Ebd4zAo//2T/6WBNbfmvtoVBjdLwa7T9daNQuNCVdA3yr8haqm2YDPj1Xj8cnpWDJrOB759BAAz7/JM4IufhA+/nsmcLsJhVDbbMBzm07ioRsHevz3nNo35moH7ZoW/P2nErvgQgh9NCq8PHMYNGolp4ToDkjwU2UjEqJUyOofi+WF+50eKwGwvPAMfjNugHnzhdHEYP/Zy6hpakdClApjB/bqcTvGevI1hg/RBT4ff/wxlEol5s6da3ebWq1GU1OT1c/a29sRERFhvr29vd3u9tjYWN7jqKtrAiOSpVyJpPsNKaYx8VFQovMq6GExAF7e+AsyEyIwtm8U5mYnu9ztE62SC9Id/X9uHIChiVG43NqJivo2fHX0Ir45/qvV89w9Ogn3jeuHwrI6vL69xC4hNUIhRV+NCqU6/kE4ACzfUYJEtQwxagV+NzoJ356ssZoNYGuesDQqOQxGE9ocTPE3d9BsDPEM+xbbdPQSfje6Lz4/dJHX/ROjwjAoSgGdrgkFJd0zKkKJUcnxxsyhuNzaCW2EEpnJ3VvVDV0mu78PR1775mqNKHf1tRgAlxra8f2RSmT3i0FBiQ7v2NQUS4hU4s95acjLEP+OsZ5+jWGx5+GO6AKfjRs3oqamBtnZ2QBgDmR27NiBZ555Bj/99JPV8WVlZUhPTwcApKeno7S01O72G2+8kfc4GAaiewOIcUzuGE0M3hFw63l1UwcOV3a3sFiQOwjDekfhrZ1lVkEA26MqNzXOqppxfWsnlu8u550s+/5/zyEhUolpQxIcBlpsE9RB2gjkpWtxY2ociiv0OFSpx9m6VhyqbIS+rdPjoIf1/BbrLcIRCilmjOiNSWlajOgTjeOXGs3nOqJPNG7/ZD/aOkM3MZj4zqWGdiRGhfG+35OTUiGVSNBl5NYvjo+bhyViTD/rL7kdnSYs313OeycY16Xa/7e/AocqGxzWx6ppNuAZBz3CjCaGU2uZQOiJ1xhPiC7w+e6776z+zW5lf/PNN1FfX4+3334b//znPzFnzhwUFxdj8+bN+PDDDwEAs2fPxvz583HzzTcjKysLa9euRV1dHaZMmeL38wh17B/3gQv1gu/KsdzVNXlwPCZdKWnv6IPErseXBB5VHq5pNritJfPuzjJU6VtxtKoREUoZ+mhUKCjl15fIFdsP75ZOE9YdugilVIqslBircy2u0HP6vUcopKJtxUHE7VhVI+djbRvmriq6wHmbPle5adbNjN8rPIO1xZUeb3/nYu+5euw957r1zrJd5bh+YC8cv9SIwrI6fHui2iqwolpC/ie6wMeV2NhYrFq1Cq+//jree+899OrVCy+88ALGjRsHABg/fjxefvllvPLKK6iurkZaWhpWrlyJmJiYwA48xHjTSoIL211dMqkEWSkxMJoYHKrQY8Wec5BIgNHJ3cGA5Rr88l2+qzxc02LAe/8957PHd2bNwUoM6x2FyYPjzT/juuWfgh7iqV0cm43ePy4FD463zoX5vFjYgpS2Fd7fKzwjmqKX1U0duHVFkdNEdeoe73+iD3zefPNNq3+PGDECn3/+udPjb7vtNtx2222+HhZxwttWEu64amHx+vYSq5yeT1ABjUqO56dmmBsdBmtNmL/sLMOkdC1kUgmMJgZ1rcF5nqTnGdMv1mop50hVg+BlCqYOjjc/h6HLhLUCB1be4rI7j7rH+48ot7OTnskfrSQctbDYcboWz2464TCRueFKEcQdJbXYf55bN/ieqL6tE0eqGlBQqkP+yiIs330m0EMiXpqTlSRI+YJAcvRFxRcFSNccrDTX1ll/9KJPl7d8hbrH+4/oZ3xIz+HLGRXLHAHL5MDz9a1YufeC2/s/v/lk0LeIKCyr87iPERGfG1Lj8NiNg7D37GUs/PqXQA/HI46+qPiqACk7Y1Kp5741nstuL3/q6VXpewoKfIhghPyjfTx3INK0EThS1QCGAbJTYjA6Jcbj/CERfbb5zMbj/m8iSnxH12zABz+cFd2yDRe2ycyWRiV1FzQV+ksSO2OSHMOtuOLsa/sgq1+MR5sdfKWnV6XvKSjwIYIR8o/2+MUmrN53wZwLsKqoQrC6PMGqlRKVg0pheR2+P10b6GHwFqGUIX94Ijq7TCiu0FvtsmRna2/KiPfJ7KSu2YDZ1/bF3wrPuJzJkQBYcOMgqJUySPMlPt2MwZVGJXeYv+iMmLfFix0FPkQwQn6Tc9QLJ9BBT7hCSsEF4UytkDosIMlFfLgCO0t6XtADAC0Go9Xyc7hCinEDeiFVG46Nx3+1+nywrYruLW2kEkq5FHOyXBc3ZQDMXFlk3vjA1vzaWaLDl0f4FWUUSkN7FwrL6zjt7HI0803b4rmj5GYiGJlUgqfy0gI9DJ9IjAzDi9MHB3oYpAdRyjz/eG0yGEWVe+KN1k4TCkp1WLn3gt2XIiFP0TKRekHuIMzNTnbZnIXd+PCPveex48rM2kSbWkDOaHzUK9BVl3gWu3PW9nfJbounBqruUeBDBJWXrnXYebmnu21kd4Vkvt2jwxVSzM1O9tGoxMtXF4aexJsCfe1dNLPISohUcmpcmj88ETtO16K4Qg+jicH8GwYiLtz9/T7ec97cef35LSfdvnc1KjmUct9cOtk8JaOJQXGFHttO1pjPB+C2c5ZL8BTq6NOJCM5y6nj/+XqsKqoI9JC8tmLPeXx97BJuyoh3f/AVj90wAHOyUyCTSjCsdxSe3xL8O8tY04cmIEIpC4rXngTWrJF9HLaEsGW5vJYQqcSskX2ga+UXfHIJVoWuOG3r/+2vwHPVJ61q/7DLWBqV3G0qARs82VWtJ2Y040N8gq2m/NCEAYgPktmfmmYD54TMWLXCHPQAgFQqCZmgBwDWHb5INUkIL7Z5uYlRYXgrfxj6xah5P1ZNs4FTsMRHfITCLzOZe8/V2xU8ZJexCjlWy6Zt8a7RjA/xqQ9+OBuSf4TP3JRmtZPF14UdxehQJfdeTkT84iMUaGw3osPom2U4dnVGAmByhhav3ToUMqkExRV6nzwfX60GY8BbvHx3sobTcbQt3jUKfIjPiKlfjj/NzU626psVzK0ySHC7f1wKBvaKgDZS2Z03s/64z5+TAfB9iQ69o89iQe4gn9X94SvQQQ/QXaHd3U44Z219yFW01EV8Qoz9cvxBAuDB8f2tEhNrmjoCPSxCPDKmXyymDU1AVkoM6lvd95sS0qcHK9FmMAblblEuydrOuFsyv21Eb6skb2KPZnyITwjdL0cukaCLEf8fMQPgpg/2oNPi5DVhssANiBAPaVRymEwMjCYGMqnE78snDIBbVuzDi9MGm3eLelJo8KaMOOws4ZYb42tsRevrB/Zy2bHdExIAUWEyq9wmT2v7BHtxRAp8iE/w6ZfDhUwqQZfRs8AnQiH16zR1p03E1yBwJ2pC/KGhvQuPrj9uvnjmpsZ5VZTRE80dRjy76QTeyh9mtVtU12xAbLgCr3x7CrUtzoOHWLUCs0b0CUjgkxgVhicmDkKsWmEXQBRX6AUNeoDuQNG26z2bFM3+/rgIheKIEobpAV+jA0Cna4JYfjMSCaDVRolqTO78u7hSFB3CI5QyfPvwOMxcWeTzbaiEBLMZ1yTi+MUGnK8X9ksNF4lRYdj4wFi7WQe2mJ8r8REKGIyMX/7+Y9UKLJw4CAlX8myczZJsO1mDF7ae8vl4WM5+fyz2GvPFnrN4xsXvk08AFQjsebhDOT7EJ2Zf29due2ogtBiM+LS4koIeQry05ZfqgAQ9wNXaNLby0rVYOnMoNCrny8m1LZ1++/uvb+tEQlQYslJiXC4N8V029LZeorPfnyWjicE7IVIckZa6iE9w6ZfjL58eoCJ6hPR0umYDjCYGhyr0OFihh0QCSCUSfHH4Ihra3S8na1RyhMmlPt8dxqV8B9+dajOvScRPZ+u9GvvOEp35uR0FZfvPXg6Z4ogU+BDBsYlxgxMiMWVwPHaW1FolOkslwN2jkxAZJse/iyvR5OMcGGosSkjPd0Hfhql/3+txs+KG9i58OHsEpFIJdM0GnKtvxT8sqj27EyaXooNDKxEuszkyqQS3j+yDjzkWWVw4MQ3jBta7XdZz5csjF/HlkYsO83WMJgY/lXHr8RUMddko8CGCcpQYFx+hwLgBvaBSyJAco8Lsa/uae92M7Bvtl9oghJCrolVyjwOIQIi22a3kKV2LATcPS0RBqQ6bjv/q8Bg2KVkTJsfBSj3C1WHoaDdYtcRwhk8NnRSOFamHJERCrZQ53dmWGBmGti4j59fTNuHZ0We2K8FQHJECHyIIo4nBqqILDj+cals6sfmXaoeJcf6uDUJIqHtwfD987eSiL1a2u5U89W5BGUprW1wuwT+ROwiTr/TkGzsgFrG9IjF6yXZOj8/W0OGyBZxrAPHExEHm/7fd2cY+T2F5He/ZoGW7ymEyMVi05STn+wRLcUQKfIjXuH5jWLarHLmpcVYfBsHw7YGQniIvPQ4SiQS1PWS5wl2VYr4aOoxu8w7/WngGk9K15s+pfWfqOCVHRyj51dAZlaRxO/PmKNBg+yB6q7qpA2/t5NdK58lJqUFRz4d2dRGvsNtJuUyTVjd12PXdGZWkQYwfGv8RQgB9a6fgzTuFJgFw16g+AIQNerhiE3iNJgb7zl7Gq5t/4XS/FoPjGjoFpY5zZwrL69wuT3EJNLzpBci1lhDbMDY3Nc6qKr3tDi+jiXF5u1jQFYd4zJM/uEWbT2DxlUqsQPe3l+F9ovHj2cu+GCIhxMKhKvE3jmXQvXU+kArL6rBww8+CFGt0NNPN5bNTo5IjNzXO5TFGE4N1h6t8ulPtvpwUPDRhAArL65C/sshpYcOeVPiQZnwIAM8idU+abzZeqcTKfgsqKNVR0EMIsRLonZifHaoSrEK1ZQ0d9nN2xZ5zbj87G9q7XNbeKSjVIX9lkceFYiMU3C7/Y/vHmnOIbMfMzmq9V3jG5e3OZr0ChWZ8iMeRujfbGl/59hTCFcPwzs5Sjx/D0i1D47HnbD30PWinCiEkNBSWdecJ8e015qj2jtHEYHXRBc5b4Z3h0sYnMTIMI/pE4/ZP9rs8zl1DakezXoFELSucEFN7CF+2rHBX8t1VifLiCj3mfXFM2AF5yDYJUuikSEIICRT2iygAvLOz1GV/MndiVHIwAKeE7WiVHHePTvI6yAKAj+4a6fPCh1xbVtCMTwjjss7sKlLnW33Ul2yDHAp6CCG+5q8+YOySkRAykzXYVcataWtje5cgQQ8grsKHlOMTwrjk6Ljq8SKTSszfQgghROxi1QrBHmv2tX2w+aFx+N3oJMEe0x+4Bj1CE1PpEgp8QhjXCNzVcXnpWlw/sJdQQyKEEMHdOaov/u83wzF3TBJUMmHyTDJTuvNuuFZgFgNfpdi4e1yxFT6kpa4QxjUCd3Uc7coihIidvq0TC776GUKWlVm85RTkUil6hQszizSsdxRO/NokyGM546uyOu4aUk8dHC+axGaAZnxCGpuj44qrSN2bwlmEEOIv35+u9clFf9muchRX6r1+nMTIMFTpW70fkBMJkUr8NrOvV4/x0IT+dtcLtrDhgtxBmJud7PS+aw5WimpLO834hDA2R8dV0pyryqGe1PEhhJBgUd3Ugc+Kq7x+nFkjewuWRGxrxjWJGD8gFm8XlHv8GIlRYbgvpx/uy+ln1ydMJpXAaGKw7VSNy8cQ05Z2CnxCnNOOv1FheHJSqs/q+BBCSDDgWmzxoQn98fWxSw4/Zzu7fFewccsv1V5Xwrb8Amy5JZ0tyLj/fD3njTK+3tLOBQU+xGnHX1eRudHEoK6VAh9CCOFiYFw4Nj2Y4/Bz1raHoVgkRobhyTzHX4C5Nqe2dOBCvdtriz9Q4EMA8Ov468kbnhBCgk2sWoF6jo0+/7r7DCalac2fs0YTYw6CYsMV0KhkaGg3un4QC/flpEAhk2KDzSySkEyM45kod4VvnflkXwU2/1wd8P5dVLnZiVCp3MyXp294QggJNktnDMXy3eWcAw+2evGO07V4a2eZVXd0pUwCg5HbB3xiVBg2PjAWMqkEO0pqsWjzSY/Gz5VlBX+jibFrVurtYwqFa+Vm2tVFOKNdXIQQ0m3pzKGYPDieVxFXXbMB7xWewaItJ62CHgCcgx7gas6N0cRg+S7Pk5ZZ4W4ali7bVQ5DlwnFFXqs3Ou+wSoXy3aVc2qG7Qu01EWcspyK1UYqYTIxtLxFCCG4WgU6L12LKYPj8f3pWrf3OVff6rLejTsSAG/MHGqeKRFiZy2X5brqpg7cuqLILljzRiCTnSnwIQ45yuOJVtHbhRBCAOD/FV3A6ZpmaFRyHLigd3t8QoQSXx72buv7bzP7YnJGvPnfQuysnT40AZ8dcj8uIYMeVqB2BtOVjNhxlsfT6ONGfIQQ0lPsPa/H3vN6zsen9FKjuMJx30Ouom16jXnT/4rdSq9RyTkFPr4QqP5dFPiEINslLMvthZTHQwghwvM26AGADUcv4r6cfubPa7b6vrvlroRIJRZOTEVsuMJh8cGESCVqmw3wZ8ZNwpUxBAIFPiHG0RJWQqTSvL2QqjETQog41bZ0WuXFcKm+//CE/rjXIliyJZNK8Gc3j+ELHV0mFJbXBWRbO+3qCiHsEpZtYFPTbMCzm06goFRH1ZgJIUTE9p+vx7aTNSiu0MPQZYJGJcfdo5MQY7MMxvbRemB8f7cFA/MytPj7PaPtenHFqoVpwOpIQ3uX+brjb1THxwkx1MxhCVHHh0vthcSoMLw0LQPz1x/3cKSEEEICJUYlx83DEpGbFserQjJ7jamuacThyqtpECP6RGPWP4pQ2yJ8YjPLsiaRt6iOD7HCZQmruqkDEsBtx3ZCCPGGLEAdC8b3j4HaTc2ankzf3oXPDlWhob3LnL9TXKE3zxC5q5vDVvCfNjQBWSkxkEklyBnQy6djZre1+xPl+IQIrktYl1s73a4ZE0KINyQSCQIxpc5nF1ZPtmxXOUxMd3FDZ/mc7vizNZG/UyyCN/QlVnqFc1ur7RWuMHdst535oTo+hBAhdAWoYm+oqG7qwKLNJ13mc7riLB/UV/y9rd3rK1lzczOUSiWUSloeETOuHzPscY46tptMDB6l/B9CCOnRlu0qR25qnMO8Gn+XNIlWyf2+rZ33jE95eTnmz58PAPj+++8xbtw43HDDDSguLhZ8cEQ49a3cktMsj5NJJRiVpIE2UgldswGH/bwOSwghwSpCKQvYc1c3dWDd4SqHuT/+Lmnyu9FJgiQ288F7xueNN95AQkICGIbBsmXLsGDBAkRERODNN9/El19+6YsxEgFwnUq0PM6fa7yEEBJKWgzGgD7/8t1nzP+fEKnEklnDkd07EoVldX4bg0Ylx305/fz2fCzeMz6nT5/GkiVLUFVVhQsXLuD3v/89fv/736O83PsOscR32AqfrkSr5DCaGBhNjN/XeAkhhARGTbMBj3x6CDtO1+LbE9V+e97np2b4fbYH8CDw6erqAsMw+Omnn3DNNdcgMjIS9fX1CAsLE2xQp06dwr333ouxY8fiuuuuwzPPPIPLly8DAI4ePYo777wTmZmZyMvLs5tl2rBhA6ZMmYJRo0bhjjvuwOHDhwUbV0/GVvh0pbG9C/PXH0f+yiK8vr3ETyMjhBAiBm/tKIPei56MXGMYtrhiIKo2Ax4EPhMmTMBjjz2GDz/8EDNmzEBFRQXmz5+PiRMnCjKg9vZ2PPDAA8jMzMSPP/6ILVu2QK/X4/nnn0dDQwMeeughzJo1CwcOHMDrr7+OpUuX4tixYwCAoqIivPbaa3jzzTdx4MAB5Ofn45FHHkFbW5sgY+vpnO3WslXTbKCGpIQQEkIYAPVedmC/YVCcy9uvH9gLH901EhsfGBuwoAfwIPB57bXXMHz4cMyZMwd/+MMf0NLSgmuuuQYvvviiIAO6ePEihgwZgvnz50OpVCI2Nha//e1vceDAAWzfvh0xMTGYM2cO5HI5xo8fj5kzZ2Lt2rUAgC+//BK33norsrKyoFAo8Kc//QmxsbHYunWrIGMLBnnpWmx6MAcfzh5B29MJIYTwFmNz7UiMCsPSGUNxsrrJ5f1KdS28Kkr7Cu8rX0REBB577DHzv4cMGYIXXnhBsAENGjQI//jHP6x+tm3bNlxzzTUoLS1FRkaG1W1paWlYv349AKCsrAy/+c1v7G4/deqUYOMLBjKpBFKphGZ1CCEkBKgVUjw0vj/+9t+zXj+WVAI8fVMa4iKUVp3euXYHsGyyGiicA5+5c+d2V9t04V//+pfXA7LEMAz++te/YteuXfj000/xr3/9C2q12uoYlUqF1tZWAEBLS4vL2/lwc6p+xY5FyDHpWihpmRBCQkGXkcHAuHAkRCq93rBiYoDF35zCX/KHIS9Diy+PXMTOEh2aO7gtk+2/UA9diwHaCCUyk4Wd/eF6jeQc+OTk5Hg6Fo80Nzdj0aJF+OWXX/Dpp59i8ODBUKvVaGqynkprb29HREQEAECtVqO9vd3u9tjYWN7PHxfnvtGZvwk5ptS+FPgQQkgo6DQxeGLDLwiTd2e3SMC9qK0jEgCvfncabZ1G8C3CvWpfhfn/+2hUeHnmMEwf3seL0fDHOfD5n//5H1+Ow8qFCxfw4IMPom/fvli/fj169epukpaRkYGffvrJ6tiysjKkp6cDANLT01FaWmp3+4033sh7DHV14urOHhcXJeiYBkUpOEX/KrkU7V0mYZ6UEEJIwHRc+SxXK6Ro7fT8c52BMHWILjW0Y96nh8yzR95ir5Xu8M7xqa+vx5o1a1BdXQ2TqfsX19nZiZKSEmzatIn/SG00NDTgj3/8I8aNG4fXX38dUunV/OspU6bg7bffxj//+U/MmTMHxcXF2Lx5Mz788EMAwOzZszF//nzcfPPNyMrKwtq1a1FXV4cpU6bwHgfDBKSHnktCjkkqkXBqRtreZcKUwfHITY3DmboWrCqqcHk8IYQQcfMm6PGFd3eV40YnLTR8QcIw/C6l8+bNw7lz59CrVy80Nzejb9+++PHHHzFnzhwsWrTI6wGtXr0ab775JtRqtV1O0eHDh3H8+HG8/vrrKCkpQa9evfDoo4/ijjvuMB+zceNG/P3vf0d1dTXS0tLwwgsv4Nprr+U9Dp1OXDM+Wm2UeUxGE2PVQ8ubLPntp2qw+BvXyd8SAO/9ZjgOXtDj/x2o9Oh5CCGEEGc+umuk10nP7LXS7XF8A5+srCxs3boV1dXVWLFiBf7v//4PGzduxJYtW7By5UqPByw2Yg18dpbYt5FIiFTiqbw0j+oi/Lu40qp0OSGEECKUmzLisLPEfRuM/71lCKYNTfDqubgGPrzr+MjlciQmJmLAgAE4ffo0AODWW2/FiROul0yI9wpKHLeRqGk24NlNJ1BQquP9mJX6dvcHEUIIEbU5WUkBbXzqzKGKRk7Hce0nKQTegU9SUhJ+/vlnREdHo6WlBZcvX0Zra6vdbioiLKOJwTsFZS6PWbar3KrLLnu/4gq9uQuvoctk9e++GuFajRBCCPG/udnJeGJiKr59eFygh2Knvq3TruChrcSoMIxK0vhpRB4kN//+97/H3Llz8c0332DGjBn44x//CLlcjjFjxvhifOSK/Wcv8yoOZTQxWFV0AZ8XV6Kx42r2vVQCq+2HVL2ZEEJ6psgwGRZPycDkwfEAALVShrnZyVhzUFy5mDcPS8Rnh6qc3v7kpFS/VnPmfdWbPXs2MjIyoNVq8fTTT2P16tVoaWnBfffd54vxkStqmrjNqOmaDSgo1eGN7SVocFCZ2bbmAlVvJoSQnmn6kHjEhiuw7WSNeaPLgtxBAIBPD1ba1epRyqQwGP2/oys3LQ5yqQRriyutrkFSCTAnK9nvfbt4JzeHCrElN5c2GHD3yn1uj31oQn+s2HPeD6MihBAiJpYbXQxdJnxxpApHKhuhVkiRFh+Ofx2o8vuX3cSoMDwxcRAWbT7p9BihOrX7bFdXXl6e09YVO3fu5PNQoia2wCe2VyTGv7HD5XJXYmQYTIwJtS3eddglhBDSc1kGEgWl9juB/WlOVhK+P13r+toVFYaND4z1ermLa+DDe6nLskEpAFy+fBn/+c9/cOedd/J9KMKDTCrBtCEJLtduZ43sjY9ptocQQkLam9+X4vqBvfDj2ctui9T62tYTNahvc/1l3N/NS3kHPrfffrvdz6ZMmYInn3wS9957ryCDIva++/mSy6BnbnYyUmLUTm8nhBASGurbOnHLx3vRXX428GPhQufHGSne29kdSUpKwrlz54R4KOKA0cTg1c2uo/btp2vRK1zhpxERQggRs4Z2o8MNLmLlzzo+vGd8Dhw4YPXvzs5OfPfddxgwYIBQYyI2Dlc24FKD611d1U0dMDIMYlRy6HvQm50QQkhwc3ddEn0dn7lz51r9WyqVIjU1FS+//LJggyLWdC3cpgAXf3OKtqcTQggRDS67ukRfx+fUKdcNLYnwtBHcpgAp6CGEECImT05KRV66FtJ8id3ussSoMPPt/sQ58Ll48aLbY/r27evVYIhjmcka9NGo3C53EUIICR4RShlaDEb3B4qURiVHbmocACAvXYvc1DgcqWqArtlgLrjoz5keFuc6PkOGDLGr38MwjNXPTp50PpXV04itjs/BX5sx79NDgR4KIYQQHxuSEInHbhyI5zafQFNHzw18AOCju0b6bZu64HV82OKEGzduRHFxMZ5++mn069cPly5dwjvvvINRo0Z5PFji3vThffCX/GF4x2aqMFolpyUuQggJIqdqmvH18V97fNAD+HebOle8Kzfn5uZi06ZN0GiuZmA3NTVh+vTp+OmnnwQfYKCIbcZHq42CTteELiNjNVVoNDGYv/54oIdICCFEQBLArtdWT9SjZ3xYLS0tMJmsm5y1trais5PaJPiDTCqxehMZTQwSIpUBK0dOCCFEeMEQ9GhUcoxK0sBoYkSR28PiHfjcdNNNePTRR7FgwQL06dMHFRUV+Nvf/oYZM2b4YnzEDZlUgqfy0gJelpwQQgix1NDehQ9+OIttp2qsvpxbNlMNBN5LXS0tLXj11Vfx3XffwWAwICwsDLfddhteeOEFKJX+q7zoa2Jd6nI2poJSHZZ+Xwo9x/LghBBCSCAtnTkUkzPiBXs8n3VnZxkMBuj1esTGxkKhCL5WCWINfGxzfCynDD/44Qz+ud95Py9CCCHE1rDECJyobvH780olwOu3DsXkwcIEP4Ln+GzZsgUzZszA119/7fSYWbNmcX044oGCEp3dri7LKcNArpkSQgjpmYb1jg5I4GNigEVbTkIqlfh12YvzjM+MGTOwZcsW5OXlOX4gicS85T0YiG3Gx10dn6UzhyI6TE47vAghhPAy85pEbP6lOmDPnxgVho0PjPX6y7vPl7qCnZgCHxPDYNYnB1xWbpZKgNduGYK/7CzrUR15CSGEBE5iZBiMJiN0rYG9bgix7Z1r4CPl+8Amkwnff/89AKC6uhpPPPEElixZgubmZv6jDGFGE4PiCj22naxBcYUeRpPzKItLd3YT092kNH94b6GHSgghJEjNGtk74EEP4N9Ch7y3s7/55pvYtm0bpkyZgpdffhnNzc3Q6/VYsmQJ/vKXv/hijEGnoFRn16zN1fY+rt3ZAWD76VosnTkUy3eVU20fQgghDrHXnM4uk/uD/UAb6b9d4bwDn8LCQnz22WdoaWnBjz/+iG+++QZxcXG46aabfDG+oFNQqnNYc6em2YBnN53AW/nD7IIfrt3ZAaC6qQOxagU2PZhjtfurvrUTy3dTMEQIIaHu4Qn9cW9OP8ikEhRX6H3yHFJJ90oEF4lRYRiVpHF/oEB4Bz719fXo27cvdu/ejYSEBPTv3x9GoxFGY8/vKeJrRhODdwvKXB6zbFc5clPjrJK8MpM16B0dhl8bOzg9j67ZYFfhGQB+vtSItcVVvMdNCCGk53hwfD+kxUfarSwkRoXhyUmpVl+uRyVpEKNWCF4DjmvQAwBPTkr1665k3oFPSkoKvv76a3z33Xe4/vrrYTKZsGrVKqSlpflifEHlSFWD2xmX6qYOHKlqsApaZFIJ7h7bD8t3lHJ6HkdThoYuEz47REEPIYQEu5F9ozFuQC/kpsa5bRUhk0pw89CEgFwfHAVi/sA78Hnuuefw7LPPQqVSYcmSJdi3bx8++eQTfPTRR74YX1Dhmrzl6LhGjtE42xvF1vqjF3lF4IQQQnqmpzf+gsmDEzC2XwwSosIwKV2LI1UNWLn3HEwmIFotR1y4EtoIJRgAaiXvfU5euXdsCnIGxAasZxfvwGfMmDEoKCgw/zsmJgb//e9/g6pdha9wTd6yPa6gRIdPfjrH6b5ZKRocqWqwekMZTQwOVTTwGishhJCeqb2LwZZfqrElgLV5XPnP0YsY0jsqYEV3ParjU15ejs8++wy//vorXnvtNXzzzTe45557fDG+gPFFHR+jiUH+yiKXy122hZy43McRNmMfgN06LyGEEBJojjbzeMNndXx++ukn3Hnnnaivr8eePXvQ3t6ODz74ACtWrPBooKGE7aTuim2SF5e8IEfYXWLPbjpBQQ8hhBDRWbar3GUNO1/hHfgsW7YMy5cvx7vvvguZTIY+ffpgxYoVWLdunS/GF3Ty0rV4K38YEmyWsxKjwhxGv/4s6kQIIYT4C7uZx9945/icP38eN954I4Du/lwAMGLECDQ0UA4JV3npWk7Z9oB/izoRQgghnrJMsXh9ewkaObRPCsSXe96BT9++fXHo0CFkZWWZf3b8+HH06dNH0IEFO0d1dhwZlaRBQqSSlqsIIYSI1oPj++H+cf3NX+CjlDI8yqFpdoW+zddDs8N7qWvevHl45JFHsHz5cnR2dmLlypWYP38+7r//fl+MLyS46tslk0owbUhCAEdHCCGEODdlcLxV0AMA1yZpwGXT1sd7zqOgVOfD0dnjPeNTXFyMl156CZs2bULfvn2xb98+LF68GNOmTfPF+IKC0cQ4XdZy17fLaGKw7VRNoIZOCCGEuPT96VocrWqwum7xqR3nqGOBL/Hezj527Fjs2bMHcjnvmKlHEWo7u6vAxsQwWLT5pNP7vpU/DBqVHPO+OOb9QAghhBAv5aVrXc7QzM1OxrZTNbzTMz66aySn9A9XuG5n5x34vPXWW2htbcXtt9+O+Ph4c4Iz0J3/EyyECHycNSRlSQC4eorEqDDMv34AXvr2tHcDIYQQQixMSovDrrI63vfTqGRoaBe+N+eSmwfj5mGJXj0G18CH97TN6tWrAcC8fV0ikYBhGEgkEpw86Xz2ItRwaUjqLq6qbupAvcCN4wghhJDiSs92Yvsi6AGAulb/beDhHfjs3LnTF+MIOp4WHrSlUckhlfDrdEsIIYS4wmWruT81tvlvPLwDn6SkJF+MI+gIVZugob2Lgh5CCCGCc5du4U9SP/ZJ9W9L1hAiROHBxKgwxKoVAoyGEEIIsSaWoAcARifH+O25KPDxEbbwoDeenJSKhKgwgUZECCGEiI9GJfd6RxcfFPj4CJeGpM5IJcDSmUORl641B1D+qW5ACCGE+NfzUzP8VsMHoMDHp5w1JNWoXKdWvX7rUEzOiAfQHUD9OS9NVFOShBBCQlt0mMzrx0iIVDpszu1rwV2FUATYhqTFFXocqtSDYYDslBg0tHdh+e5yq51fiVFheHJSqtdvggiFFC2dJm+HTgghhDhk6PL+GmMyBeY6xbuAYagQqnIz4Lh6c3SYDL8dnYTMJA0ut3Y67dBuNDHIX1lETUoJIYQEJaFmfbgWMKSlLh9jqzfbBi6NHUas3HsBi7achEIuRVZKjMM1TqHqARFCCAler88YgvvHpQR6GB5Ztqvcqjm3r1Hg40Ncqjc3tHfh2U0nzL1PbDu11zR1cHquCIX3662EEEJ6Fo1Kjrfyh2Hq4ASM6Rcb6OEAAO/NONVNHThS5VklaU9Qjo8P8ZmtWbarHCaGwfJd1nk/MW4SoVl3ZfbF6v0VHo2TEEKIbzwwLgVri6vQ5qO8ywaLCszsLuBArxJ4MncjVNFfLmjGx4f4vJDVTR1YtPmk3RtWz6GsuFQCRAqQYU8IIcQz4Qrryym7YykzOcZnQQ/rje0l5qWia5M0Pn0uXxGi6C9XQTnjU1dXhxdffBH79++HTCZDfn4+nn32Wcjl/j1df72QJgZ4/4dzfnkuQggh9lodBDc/X2zExuOXfP7cDe1d+GTveaw7XIXGDt80EfVUjFoBvZtm24lRYRjlx4AtKGd8nnjiCYSHh+OHH37A+vXrsXfvXvzzn//0+ziEqN5MCCGk56lpNmDNwUqvAhE+uTIr910QPOiJVskxNzvZq8d4cuIgPDShv+tjJqX6tYBh0M34nD9/Hvv378d///tfqNVqpKSk4NFHH8Xbb7+NBx54gPPjtLS0Qoid/vPH98WL35zy+nEIIYSEjhi1Al/8KQtPb/wFRy82BWQML996DbJSYpAWK8d7heWobXY9c+NIhLQLvx+pRVKExO4xEqKUeOzGQcjpq0ZLS4vX45VIJJy2swdd4FNaWoqYmBgkJiaaf5aamoqLFy+isbER0dHRnB5nxIh0NDUF5s1GCCEktFUAGPpGYMcwe7lvH6MCwJ+8fwqzqKgoNDY2uj0u6Ja6WlpaoFarrX7G/ru1tTUQQyKEEEKISATdjE94eDja2tqsfsb+OyIigvPj/PxzqSBLXbaKL+jxxIZfeN1HAuC+6/vjy4NVaOSwy4sQQgg/KrkEcqkUzQb/JgdrIxR4YWoGLrd1okrfho3Hf4Wuhf+SklDiIxV4PDcVuWlxLo/7/nQtlnxXwvlxE6KU+OJP2T7N5ZFIuD120AU+6enp0Ov10Ol00Gq7S2CXl5ejd+/eiIpyv/bHCg8PF6xlhaVWpgVSpYrz8YlRYXhqUiqio1VYvb8aUmXQvWSEEBJwhiv/SZUKvz7v5U7g5OVOpMVHYnXBBQAySJWBK08iVSoRpg5HeLjriYJkbSeva5muAyip70JWSoyXI3SOY9wTfIHPgAEDkJWVhTfeeANLlixBfX09PvzwQ8yePTvQQwPAfYv7/eNSMKZfLEYlaSCRALM+OeDjkRFCCAmElXsv2NUBCpTaZgOe3XTCbf8sT4ol7j9fD12zwWlvSn8JyialOp0OS5YsQVFREaRSKWbNmoU///nPkMm4R9FCNim1xKXpaGJUGDY+MNb8piiu0GPeF8eEHwwhhBDigO11yBG2F6UnEiKVeCovTZDmpCyuTUqDMvARgq8CH8D9m8U20t52qgYv0JZ4QgghNqSS7iK2vnBfTgrG9o+1mp0xmhgcqWowz9zUt3Zi+e5yj9tkCNWZHaDAx2u+DHyA7uDn3YIyqzdLYlQYnpyUavcmoBkfQgghjszNTsaag5U+fQ52dgaA3XUrIVKJhZNScbauFSv2nOf92FxmlriiwMdLvg58APvI2dmap4lhMOuTA7jU0O7bARFCCOkx5mYnY0HuIIdfpP3toQn9MTAu3K7RNhcf3TVSkKRnCny85I/AhyuJBDj4azPmfXoo0EMhhBAiEpazJewX6ZqmDizbVc6pwbXQEiKVWDgxFbHhCuiaDTh7uQWf7Ktwe7//vWUIpg1N8Pr5uQY+4kgjJ25NH94Hf8kfZtf7KyFSKZrdAIQQQvynuqkDR6oaAAAyqQRZKTG4eVgiFk3NCMh4apoNWLTlJBrauzBtaALG9IvldD9/dmYHgnA7ezDLy9DixtQ4u8SyRVtOBnpohBBCAkDnYFkpL12Lt/KH2S1/xaoVqHfTKV0Iy3aVIzc1jtOWd393Zgco8BEl29yfzOSrbwo2qmePy19ZFKBREkIICTRnsyV56Vrk2nxRHtEnGrd/st/nuUDsTFRWSgyeyktzuYvZ353ZAQp8RMdRklpCpBJLZg1Hdu9Iq2OPVDUENJmNEEJI4MSqFRjRx3njbcsvyix3gYhQ2JmovHQt5mYnY21xpdW2e6kEmJOVLGgdH64oOURE2Po+tsFMTbMBj3x6CAUlOqufO5riJIQQEhrq2zpx+yf7UVCqc3mc0cSguEKPbSdroFHJccOgXj4fGzsTVVCqw5qDlXa1hkwMsOZgpdux+wLN+IiE0cTg3YIyl8e8u6scN6bGmacF/Z0QRgghRFxq3LSYEHKru0Ylh0IK6Fpd7xhj83a4XNfYfCB/LnfRjI9IuFu2YmCdwQ9c7ZVCCCEkOMWqFXhpWjoi3DQuXbarHEabaRVnqwieen5qBrY8PB4PT+jv8jg2b4dLOobtdc0fKPARCa7LVuxxbAL0TRnxvhwWIYSQAHpuSjr6atRoMRhdHmcbQBi6TFi6vUSQMSRGhZlnlGRSCR4Y3x9vOSivYnkcwP+65i+01CUSXJettJFKUVTpJIQQ4lsalRy5qXHYcbqW0/EHLnR3P6/Qt2Hd4YteFzG8f1wKxvSLddhVwNGuMdvj+FzX/IkCH5FwV+9AAiAhKgz1bZ1YtJnq9hBCSLBraO/CkaoGzoEBlyrJXMVHKPDg+AEuc28c7RqzJNY6PrTUJRIyqcTcBM4RBsCUwVos31Xuv0ERQggJKF2zISD5nH++Kd3rhGN31zUgMHV8KPAREbbegTOfHqyi5S1CCAkh2kglpwBCKOEKqdMdYp5gq0i7ywfyJ2pS6kQgmpSylZgpuCGEEGFI0D1j3hNZNiEFhN2aDgCvzxiCmsYOHL3YgJhIFSan9kJ2v1jz89l2EXCU68OVkI/lDNcmpZTjIyJUiZkQQoQTq1Zg4wNjseH4JSzffSbQw+HN0TKQUHMVd49OwtTB3R3R50pSoNVGWX3hd9ZF4Km8NI9madzlA/kTLXWJCFViJoQQ4Tw3JR1qpQy/zUxymyMToZQhKsx1rRxfsZ33iFUr8OD4fujsMqG4Qg+jiTHX5KltEabJaJTK+byHqy4Cz246EZBqy0KiGR8RqdC3ef0YYXIpOrpMAoyGEEJ6psSoMDw5KRW5qXEortBD12zA7SP74OM9553ep8VgtAtAXIlQyDBhYCx2lOi8Xkpj0J1b09rZ/dld39aJlXsvmG+Pj1DAYBR2wW7FnvNI1UbYzd54Um3ZH8tYQqLARySMJgYbjl1ye1xCpBILJ6Zi+e5yq2hco5Kjob2Lgh5CSEibPlSLCQPiUK5rwTs7S61mSMIVUihkUjQ4qW/DJ7Ro6TQiWiWHRAJO+aBKmcRl8MIGPY4INctjiw1g5LKrQQqfastZKTGCL4n5AwU+IsE1v+favtGYPDgek9K15gg7NlyBV7495YdREkKIuH13UofvTjpeimntNAGdJkglsGua6Yn/HPuV87FCz9gIgQ1gsvvFmH/GNeWisKwODe1dDju9u+sfFmgU+IgE1zfb9yU6TC7VIS9da04UK67Q++wbASGEBBshgp5gUdPUgYMX9OiobESYyYTYcAWn+317oho7S1xXlA5EA1IuKPARCT4lu23fTJQUTQghxBPLdpVbtbaIj1AgQilz2xuMSzsMyyUxMaFdXSIxKkmD+AhukbZtMzp/9zkhhBAifhoXO7dYtgFMbUun26CHDzF+MafARyRkUgluv7Yv5+Mt30z1rbTMRQghwU6jkvNqXXHXKO7XFF8R4xdzCnxEpF+MmvOx7JvJaGKwfDf17yKEkGD3/NQMbHowB/ePS+F0fP9e4Q7bRcSqua0uuJIQqXQbhAWiASkXlOMjIlwj41i1wvxmomrPhBAS3Ni6ROwOqTH9Yjl1YtdGKpGVEoPc1DirOjs1TR146dvTXo2J7R3maFcXKxANSLmgwEdE2A687gKZZyanUWIzIYQEOY1KjqUzhmJ0SoxVAMHlWmE522LbLqK4Qs/p+R+e0B8bjl2yeh7bIOyt/GF2dXxsjxEbCnxEhO3A6yqCDldIIZVc/QMQ4/opIYQQ7z0/NQNj+sfa/ZzLtWLWiN7YcbrWYSXlUUkac9FbZxKjwnBvTj/cm9PPZVXmvHSt3YyS2Cs3U3d2JwLRnZ1VUKrDG9tLXL4p2cJQ1NGdEEKCS7RKjsVTM8wzJs5aQji7Vth2pI8Ok+F3Wcm4L6ef+X6ugiYAWDpjKGLDFT0mmAG4d2enwMeJQAY+RhODmSv2uSxKmBgVho0PjOX8JiaEENIzfDh7hHmmx1VLCMB1jo0tjUqO5yan27U8shWukCIyTN6j2lAAFPh4LZCBT3GFHvO+OOb2uI/uGmlet32v8AzWHKz08cgIIYR4IlatwOxRfbDuUBUaO5zXyeHzpTZaJUcjh0KCQhJrGwqAe+BD29lFiGvCMnuc0cRg08/ce8YQQgjxr06jESv3XnAZ9ABXd0Jx6ZLu76AH6K70bOzhPT8ouVmEuCYs17UasO1kDWqbO1zmAxFCCAmsZoPz7uuA/U4osZYqEWsbCj4o8BEhLlsVpRJg+e4zfhwVIYQQoUUoZXhr5lBk94u1Sh4Wc6kSd2NzlowtFhT4iBCXrYo9fKaREEIIgBaDEYu/OYnfjk5Gvxi1OVAQc6kSV2NzlYwtltwgSm52IpDJzSxHbyCphIIeQgjxpesGxGLPuXoE6qM2IVKJhZNSsXyX+91XrZ2ul9CEZpl8bctdMravE6NpV5eXxBD4AFenDDukUpz7tZGWtwghxMcSo8Iw85oE/INDWwhfmpud7HK3bphcio4u/wY+joIXo4lBcYUeizaf4LxjzRdoV1cPwb5htp2sQXGF3i5bXiaVILtfDG4blYS4CPFOfRJCSLCobupAZnIMolWBzQbZ9POvWDpjqNNx+DLoCVdYhweJUWEOg56CUh3yVxZh/vrjbnessYnRgUY5PgHEdy1US4EPIYT4RV2LAc9PScdzm08GbAwN7V0or2uBSi5Fo4ePIQHw8vQMXGzswIo95znf7638YVDIpC4TlD0pniuGpG0KfALE2RumptmAZzedcBhZZyZrEKNWQN/mvKIzIYQQ7+07dxkHKwI/O7H2QAXaujzPu2AAXKhvw9j+sVg6cyiWFZS57AoAdFd4HmOzywyw3q3VK1zhts6QI2JI2qbAJwC4FKZatqscualxdm+8rJRo7Cyp8+XwCCEk5G09WRvoIQCAV0EPa1VRBVYVVSAhUoknJ6Xh3OVWfOxi9mfG8ES7a8+Oklq89X0p9F7UjLPsGB9IlOMTAFwKU9muhX738yXMXFFEQQ8hhPQwmgDnCrFqmg1YtOUkBmkjMDc72elxaw9W4b3CM+Yc1IVf/YxFm096FfQAV6tSB5o4Xo0Qw7clRUFJ97KYCDaZEUII4UijkuP5qRn4+WKjV70UhU5xWFZQDhPjOjF6zcFKbPz5V0HaYthWpQ40CnwCgOsapzZSCaOJwTsFZRT0EEKISEUopGixqKcTrZLjd6OTcF9OPxSW13kV9CREKnHbiN5YufeCEEMFAFQ3d3A6ztugR6OSY+mMoRidEiOKmR4WBT4BwKUlBbsWKtZ+LYQQQrrJZFJ8cNs1qGsxoL6tE7FqBbQRShy4UI/Xt5326rGfyktDp59r9Qjl+akZGNM/NtDDsEOBTwBwaUnBroWKYesfIYQQ5xrbu/DV0Ys4fqlJsC+q7DJZXroWxRV6QR7Tn8SS1+QIVW52wh+Vmx3V8bFdCy2u0GPeF8d8OxBCCCEBd/PQBPSJDkN2SozV8pChy4RbP97ndXIxy5+tLnzdpsIStazwkr9aVrjrYms0MchfWYTaZgPl+RBCSJBy1s7B0Rdkb0kAv11PfN2mwhK1rOghZFIJslJiMG1oArIcJIDJpBL8OS/N48ePVSu8HSIhhBAfeyJ3kNPKyEIGPWEyiV+/RLOlWdy1Z/InmvFxQixNSoHuKPbgr8146eufeXdqXzpjKJbvdt3hlxBCSGDZtitiZ/vF+tkdq1Zg2pB4fH74ottjrx/YCyW1zZzbM3mKlrq8JLbAR6uNQnVNIw5XXl0WG9EnGscvNWJ3qQ5bTlSj2aJBnGWukCf9VAghhPjf0plDEatWYP/5eqwqClx3+KzkaByuarT6ci0BcN3AXrhnTLJ517G3OahC5gD12MCnsrISb775Jg4ePAiGYZCVlYVFixYhJSUFAHD27Fm88sorOHbsGCIiInDPPfdg3rx55vsXFhbinXfeQUVFBfr06YNnnnkGkyZN4j0OMQY+rsbkLlfIF+vEhBBChMVlJt/X3OUAscGKELNSQuYA9dgcn/nz50Oj0aCgoAAFBQWIiYnBo48+CgDo7OzEvHnzMGLECBQVFWHFihVYu3Ytvv32WwDAuXPn8Nhjj+Hxxx/HwYMH8dhjj+GJJ55AdXV1IE/JL9zlCuWla7HpwRx8dNdI3D8uJUCjJIQQ4kqggx7AfeLzsl3lMJoYc2kWb9i2Z/IHUQU+DQ0N0Gq1ePzxxxEeHo6IiAj84Q9/QElJCRoaGnDgwAHU1NRgwYIFUCqVGDZsGObOnYu1a9cCADZs2IDs7GxMnjwZcrkct9xyC8aMGYN169YF+MzEgQ2OBvaKCPRQCCFElCQAfpvZFx/dNRJFT96Aoy9NxRBtWKCH5ROezrFYBit56VrcPTrJq3H4u16d3ysMtbe3O52BiY+PxyeffGL1s23btiEpKQkajQalpaUYOHAglMqrLR/S0tKwYsUKAEBZWRkyMjKs7p+WloZTp07xHqdEPNW1zWMRakxcW2YQQkio+X1WEhZOSgXQ/Zm794wOl9tFMA0jMI1Kji0P5eA/Ry/i45/O8e4Cr2sxmK9JuWlx+OxQlcdj0UYqBbm+cX0Mvwc+R48exR/+8AeHt33wwQeYPHmy+d+fffYZVq1ahb///e8AgJaWFqjVaqv7qNVqtLa2Or1dpVKZb+cjLs79OqG/CTWmKb0i0WdbCX5taKfaQIQQYuHfxVW4YWgipg/vg+9+voRHPj3U4z4n+2hUaGjrRKvB6PQYqVSCvr01yGkz4q+FZ3k/R2rfGHM+jafXFAmA3hoVpoxK9msvL78HPjk5OTh92nXvEoPBgKVLl2Lr1q34+OOPMW7cOABAeHg42trarI5ta2tDRET30o1arUZ7e7vV7e3t7ebb+airE1dyc1xclKBjWpg7CM/QTi9CCLHCAHhm/TF0tRuw5LuSHhf0xKoVWDwlDf+z/meXx9W3dmLboQocrNTzfo7EqDAMilJAp2sy/8yTawpz5X71l5t5j8ER9lrpjqhyfADg8uXLmDt3Lo4cOYL169ebgx4ASE9Px7lz59DVdbVsd1lZGdLT0wEAGRkZKC0ttXo8y9v5YBhx/Sf0mCala/FW/jDEiLifCiGEBEJjexf+Z/3PPXIXbH1bJw5XcksWfm7LSazax3/L/JOTUiGVSBxeUxJsUikSo8IwNzvZ4c/fyh+GSelawa+V7ojqqtfZ2YkHHngAsbGx+OCDD6BSqaxuz8nJQWxsLN5991088cQTOHv2LNasWYOFCxcCAPLz87F69Wps3boVU6dOxfbt27F//34sXrw4EKcjennpWlw/sBduXVEEfVtnoIdDCCFEAFyrIjfy7P1l20vSVl66FrmpcQ5Lq8y/YaDLkiv+JKo6Ptu3b8djjz2GsLAwyGQyq9u++eYb9O3bF+fPn8eSJUtw9OhRhIeH45577sFDDz1kPu6HH37AO++8gwsXLiApKQlPP/00cnNzeY+lp9Xx8QYVOCSEkOARo1YI+mX27tFJyE2LC2iwwkWPLWAoFqEU+ABU4JAQQoi16DAZfpeVjPty+ok64GFR4OOlUAt8AOvqz7HhCrzy7SnUttASGCGEBIuoMBmaOpzv9nLEF321fKHHVm4mgWNZ/Xls/1j8+Sb+SeFCiqbEa0IIEdR9Of1436em2YBnN51AQanOByPyPwp8iFN5brL0w5UyJ/f0XqxagddvHeKzxyeEEDF4K38Ytj08Dgo/rSS9/9+z0Kg8++xmW1X0dPSVmrjkKkt/3nUDcMtHe9HAc9qUi/q2TkglEiREKinviBASdGx3SO158kY0t3fhlo/3oI3fZiteTAAa2j37zGZbVWSlxAg6Jn+jwIe4xS6B2VLKpXh+2mCf7Qirb+3EU3lptOOMEBI07stJwdj+seYvkIYuE9YfvYhKfTuSY1TYMf96GLpMeHT9MZysFqawn5Ac9dWyzA8N9FZ1LijwIV5hl8N8sSOM/QPSqORo4FlvghBCxGhQXIT5i+R7hWewtrjSqiP73wrPYE5WMv51z2gYTQx+OqvD01+fhCkww7XD9npkg53Csjp8e7LGavu82JOhKfAhXnO0HFbf2ok3d5TaBSwalRzPTUnH8l3lTgMlCYCEqDCMStLgSFUDBT2EkKARG66A0cTgxa2n8P3pWrvbTQyw5mAlAGBB7iB0mSSiCXoSr3wuuyt/wiZDv5U/TJTBDwU+RBCOlsMmpWtRXKHHoUo9GAbITonB6JQYyKQSSCUSl0tYT01KhUwqcTitSgghPdXhqga8vPUkdK2uv9B9erAS2f00eGdnqcvj/OnJSakoLK/jnH6wbFc5clPjRLfsRXV8nAjFOj7+5uhbQ2JUGF697Rpk944EwwDFFXrM++JYAEdJCCEEABRSIFIpRX079zmoj+4a6bdkaK51fGjGhwSMoyWyzGQNEhOizV1/RyVpfLKzSwL0uK7LhBASSJ0m8Ap6AMfJ0IFGdXxIQFkWTcy6sgxme/tTeWmCPmesWsEp6IlUynDv2BRBn5sQQkJJXatBdLV/KPAhoueskKLGw8rO04cmcDpu5vDeSNVGePQchBASSDFK98f4w/LdZ5C/skhUVZ9pqYv0CM4KKRaW1znME5o6OB7bTtXY/fzJSanQqOT47FCV2+fMTYvzybkQQoiv6a989C2ekobXvy8L6FjEtsuLAh/SYzjaOeaqsvT8GwY6/LnRxLjNG2K3bQKg6tGEkB7rH/sqsOTmwXjp29OBHopodnnRUhfp8ZzlCbn6ubu8oSevbKf3RY4RIYTwlRgZhg9nj8D94/jlHVY3daDeorhgILEtLwKNAh8Sklw1YLWdjnXXrJUQQnztybxUjOkfi/ty+iNGreB131i1wu7zK1Ae/88xLPzqGJoDWJiW6vg4IaaaOcFax8cRf58rnx4zzo51V8XUU2q5BG1dQf6CE+KhUClJIZUA9+akgGGAc5fbUFxRz7vJ6MKJg5AQGYZFW046Peaje0ZjSIwKef+3x2+VojO0aqz94xjBHo9rHR8KfJwQU5BBgY/4WQZFvcK7t8vXt3bigr4NG45eRG3L1anmWLWC09TzQxP6Y8We8z4cNSE9y4Pj+qF/r3BoI5VYd6gKu8rq/PK8XP9mxSwhUolpQxIcbvp4alIq7powEDpdEw5e8G/RWIVUgj0LbxDksaiAISF+5KyDPQDcl9PPaqZoRJ9o3P7JftQ2G5x+Y02MCsN9Of2Qqo2wm02KVStwyzUJWHvQ/c40QoJJWkKkeRn6wIV6nwY+s6/tg1HJGqT2jUH/CDlm/WO/R7O6c7OTHQYbT0wchNKaZqwqqhBy2E7VNBuw5mAlls4cili1wmrmWi67Osvtq6KxznSaGNy2ch82PjjOL88HUOBDiM85Cor+nJfmst8Nm1ztatfayL4anyyxESJWlruCundd+iZokEqAiWlx0F/JQ5FJJVg4KRWLNjtfKrLFls/IS9c63WEaq1b4LfBh/XX3GWx8YKzTJX12QwfXflxCuNhoQENrJzTh/HKXPEVLXU6Iaamlpy7/eCJUzlUiAQ7+2oyXvv7ZYa0hrrUu2CW21UUXUHRe76PREiIebO+n/efrMX/9cZ88h1ohRVvn1UyXhEglZo3sw2np+b6cFIztH+syX5BlNDHIX1kk2JeXsf002H/B/a4p2/5Zjj53C0p1eH17CRr9lIQ8sm8UPrk706vHoKUuQkRu+vA+yEyIwOFKbsnVjrCzSadrminwIYISa/Iw2/upvtW7nJvrB/bCnnOXYdlNgT1ny6AH6F4m4ppvNyguwmVTTttNEgsnprpMOuaDS9ADcOuflZsah3flUjS6OEYhlaBToHYUvzZ1CPI4XFDgQ0gAucoN4mP2tX3xt8Iz8FVLnGiV3G/f/Ih/aFRyNLh4TcUY9ADABX0bAEDr5fbse8Yk4638YVh/9CIq9e3oGx2GtQcroGv17n1e12rAtpM1Dr/IONoBmhCpRG5qHH44U+ezv19bXH53R6oa3M5EdZoYLJw4CMt3n/F6TL2jwrx+DK4o8CEkCCjlUszJSsaag5WCPi47bW80MT5bViD+93juQNw9Ohm7SnWCzTb4Czvz8scxKR4n4bKV2WVSCX6f1V2Lq7hC73XQI5XAKghIiFTiqbw05KVrUVCqc5g3U9NsQE2zf3anAdZV6V0p5Jg4ztYI8na5btltw726Px9UwJCQILEgdxDmZieDy0pZdJjM7TGJkWF4aMIAZKXEICslBmoFfVwEgzCZBI1tXThUocfOktpAD8cjK/acx+2f7Me0IdwaDttiNw9Y4rL8447tjA3bo2rH6Vq8WxDYflmsqYPj3S6nF5TqOPUzBICEqDCvq9sna1R+S2wGKPAhJKgsyB2EHxZcj4UTB2H8gFhEKK0DHLYy9eJpg90+1pN5qVZtPm7KiPfJmIl/dRgZrN5fgUfXH8eOEvF0zOaL3Z49NzuZc1ViR5XZWVyXzh6e0N/u+dx92fjLTvHsvtx+uhZGF2tqRhPDOUhjZ49cVbd/9eYMXDcgxuljJGtU2PDAWE7PJxTa1eWEmHYVhcpOJyB0ztVf5+mqMrWzXRsalRzPT82wuzgYuky4/m8/ep37MSZFg6MXG2EwBvELTPwmMSoMX903BscvNaKwrA7fnqg2b0MHupdipg9NQG5anNvK7O52WCVGhWHjlYs0+3dV12oQJMfFn1zt6uJTwNA2iHRXCb+htRNPbvwZvzZ1oHdUGJbdNlzQmR7a1UUIcZk8zdYIOlShx8EKPSQSYHRyjFVDV0tKuRT3ZHuXRySVAH+9YwR+PHvZr3VCvCXWHU6ku/Hl+qMX8dvMJGSlxODx3EGc29BY4lK/xnKJjP272nayRpDz8CdXy3pcl/zuHp1k9+XI3WYNTbjC6y3rQqDAh5AQJpNKMKZ/LMb0j+V0/ILcQQCAtcWVdtuAfzu6L8pqW3CwwvmW2jlZyVDKpeapcdsdLkqZRJQzQeIbEbG0fPcZrD1YaU4k5rNT0naWYumMoVi+u5xzfS1vd5cFgqsxcz2f3LQ4oYbjdxT4EEJ4WZA7CPOuG2DeBpwco8Lsa/tCKe9OGXyv8IxdYCSVdAc9bOAEwGlV6sLyOqpITXhjE4md5fA44mx7+cJJqd1tHVoMSO0bg0FRCkgljmeNRiVp3JYG8KUYldxqac/d7GTClb8zZ7i0rOC6M0ysKMfHCTHlmIRK3gsQOuca7Odp6DJZBUaPTB6MRn0L53Nlv4XvP1/v95L+pGdj83C47Fxytaz1Vv4w3JShdfh3ajlLdEHfFrBmwmx+0/87UIHPiyvR2OG+a7ujHD7bzyMuvxuuwaU/UY4PISRglHKpuT6KRALzbBBXbK7AqCQNtvxSHfDZn8TIMJgYE2pbenaH7lBQ3dSBI1UNbqsnv7Oz1OXjLNtVjokOlnMczRIFypOTUvHj2cu8Aq+G9i63M2POlqL5ttQRKwp8CCGiFYiGiY48mZcKAH4Zh1IqgUImQYtN24SeZkSfKIwf2CsgsyHuEnRXF11wG8RWN3XgcGUDpidEm3/mbibEX9gAJDc1DjNX7PPoMSwbvjriqkFyT0eBDyFE1Jx9+3QkQilDi8H9dD9Xtt9wuY7DGzOHJ+LJSWm4dUUR9G3imGGKUEjx7OR0aCOUeOW7027PXy2XYuXvRkEmlaDNYMSnByv9miDuKkG3oFSHjzkGY7qWq+fJp76Nr9w/LgVj+l1tgPqPvec9noXkMjMmVEsdsaHAhxAiepbfPgvL6vDtyRqroECjkuN3o5PwhzEpuP2T/V4FJrFqBRZOHIQEi7YGjsZR09SBZbvPuA1O1AopJABaOc7gjOgbDaVcikVT0kUxuwAALZ0mJESFISslBn/OS8Mzbsb1yi1DIJNKUFCqE7yNiju2ibdGE2Mu2cCAwYZj1ZwfSxtxNYDi0rvKlxKjwvDg+AFWdbi4BnDOCFGtuieiwIcQ0iOw3z7d1WpxtzQ2JysJa4udl+N/bkq6yxwGy2/BVQ3tbi8+bZ0mfDh7BL49WYPNv7i/6PaOVgHoDrLGD4jF3nP1bu/jD+xFMi9Di4/uGY1n1x+z28nEJs5eP7AX1h6owEd7zvl9nJa1dpwV6eQiMSoMmclXA6hABwmW5yXU7JOrmTF3xQh7Mgp8CCE9jrvCjO4SM0cmaQRJ3EyJUXM67nJrJxZPzUBheZ3Li7DtbMU4EQU+lhfJ6cP7IDMhAgcv6HGoUg+GAbJTYjA6JQYf/HAWizaf8FuncZbt6+dtPo5tPy9/1OvRqOQIk0vdvi+FmH1ytSXd2TZ/tk5ST0eBDyEk6LhLzBQqcZPrxVAbqYRMKsHiqRmcKwMDwOxr++JvhWcECSLuHt0X6w5f9OixHF0kZVIJxvaPxViL4pfvFZ7x+dJWZlI0zuha0GCxdTtGJccTEweZL8pcdm258tCE/nYXeC71bbz1/NQMTu9LIWafHDVqBZwHjJ7USRIralJKCAlK7KzQtKEJDttwuLudC/Zi6Ipl0OCqmaOjC4pSLsWcK2UBvFVQWofXbhni0X2dXSQtGbpMWFvsedAjl3YX33NFo5LjcFWjVdADAPr2LizafBIFpd1NV49UNXic9JsYGYb7cvrZ/ZzdYegLCZFKLJ05FBqVHDtO1wIAJg+Od/q+9Gb2yVWjVi5LaMt2lbtsctoT0IwPIYR4iG9/J4D/bJOzNiF8VTd1oFzXgocn9MeGY5esZi40Kjnau0zo6LJOwHbWsNaR9Uc9m01idZmAyRlalx3jO42uE8TZLdo1TR0ej+PJPOdBntNl1MgwtHUZPcolenhCfwzoFY7lu8o5Ly15Mvt056i+uClD6/K9drjS/RIal91ggLhzhCjwIYQQL3hS7I3vNmHLNiH7ztXj+KVGNFvMenBtmcBWwY6PUOChCf3RL0btsvKwq8c0mhirHJ9yXQvn83Fm/3nX+UzudsZVN3VgddEFrDvkPHndGdvXi71wd1Q2IsxkMl+4XbVacRUA275G7PMBjutDuVpa8qS+VWy4wu17znL7vsvj3ARHYs8RopYVToiplUCwtzewFCrnGirnCYTOuZoYBmeaOlF+UQ9thG+/4dp+mzaZGDy6/jjvx1k6c6jdTIMt2xYQEglw8Ndmh7u6eqLoMDmWzhxqtazk6YXb0f0siw3aBksAkL+yiNfv3/b5/nfbaTRxaFUBOG81wf6NfneoAvO+OOb2cT66a6TTICqQ7S6oZQUhhPiRTCrB+NQ4pGuUPg/wbGeMjCbGo8Tbv+woQ72bOkS2SxsFJTq3dXw8oVZI0RaAatWNHV2QSSVWQY+nyb3uljFtg4XiCr1XS0vs87249RS+v5Ib5Iq7as2Zyd41KOWaI+RqDP5Ayc2EENLDeZp46y7oYRWW1QG4slvKR9WLb8qI98njcsEu3QiR3MsnaZ7r7ixXx8mkErwxYygeHG+fkG2LDaJcPZa795GrRHcu2+zdjcEfKPAhhJAg4GzHmBC+O1ljXl7zxXZujUqORZPT3Y5do/LNIgW7S4rrhbu4Qi/o8wpxXP/YcE6P5S7Y4rvz0BIbIHs7Bl+jpS5CCAkSlkstBy7U45N9FW7vExkms0qUdqS+rdO8fMPVzUPjEaHsvsRIJcAXRy45Pfb5qRlQyqVuE3ZvGZaIKJUcG45etNqunhgVhpF9ozkt99iyXLrhen6LNp/A4mmDvc5V4bI7KzEqDCP6RKO4Qu9yh5SQQZQnda4KSnX4jGNSuT+KQbpCgQ8hhAQRdqllVJIGm3+udntRnZQWh88PX3T7uOwFkKvbRvSxykvJ6hfrduebsx1yUglgYmC+sCZEKvHwhP5IiVFDG6lEXYsBi785xXlsliyXbrieX2OHUZBiflx2Z00dHG/Xf85RojXXIMpZfo6jsXHdecinhQafMfgKLXURQkgQ4pqvMZHjhZv91s9lKS3BYtcSKy9di00P5uCju0bif28Zgo/uGomND4y1Cxwsj7t7dBIA2NUHqmk24OM956GQS1Hf2okXt3oW9DxsU6GZ6/mxhCjm52ppaW52MtYcrLQLZthEa7ZgI+B9fo43+CyB+moMfFDgQwghQYpLvgaf6tMyqQR/5pBE/VRemsOLG9fEX5lUglFJGuwscb109cb2EizactKjwokJkUrca1OhWSaVYNqQBM6PIVSibm5qHF6ZPhh/GpuMW4bG496xKXhhajq+O+m6qa1t4OVNfo43uC4R3j06SRR1fGipixBCgpi7fA2+1ae5dGcX4uLGZRbBmzpCjoIzo4nBxp9/5fU43ibqOqr9AwCr97vPz3K01V2oPnR8cF0izE2L89kY+KDAhxBCgpy7fA2+1adddWcX6gLry50/znaHrS66wLvthDeJut52kAcc/574VgZ3hmvbCaHzi3yNAh9CCCG8ZwocdWcXki93/jS0d9klJxtNDOddSSz2Yu5JXyo+CcGu+Or3xKd6tSc96wJJ1Dk+Tz/9NObOnWv1s7Nnz+KPf/wjMjMzcf311+Ojjz6yur2wsBAzZ87EqFGjcPPNN2PXrl3+HDIhhPRYQnSsFwrfRGNPWObIHKlq4D3b8+SkVBSW1yF/ZRHmfXEML2w9hXlfHEP+yiKrxGNHhKiJ5KtZFHYmiktSNStQ+UWeEG3gs379emzZssXqZ52dnZg3bx5GjBiBoqIirFixAmvXrsW3334LADh37hwee+wxPP744zh48CAee+wxPPHEE6iudp0gRgghRFy47FKK9rKgoWVyMp+ltcTI7os5AN4BAkuIpTxfzKJ4U72a6869QBNl4FNWVoYPP/wQd955p9XPDxw4gJqaGixYsABKpRLDhg3D3LlzsXbtWgDAhg0bkJ2djcmTJ0Mul+OWW27BmDFjsG7dukCcBiGEEC+4m0VYPDXD6+dgAxCuS0YzrknExgfHIjc1zqv2Ft4sUflyFsXbthNimjV0xu85Pu3t7U5nYOLj4yGVSrFw4UK8/PLLOHbsGM6ePWu+vbS0FAMHDoRSefUNk5aWhhUrVgDoDpgyMqz/ENLS0nDqFP8aDxIRvVbsWMQ0Jl8JlXMNlfMEQudc6Tx946YMLSamxeFwZQN0LQZoI5TITL6aQ/OX/GF4x0FS9qwRvfHxnvNuH18bqYREwq1Bp0Ylx4vTMiCTSnDwAvcGo9n9Yuxu4/J8lufz8rQMXG7rtDt/IVi+proWjv3DWgyie69zHY/fA5+jR4/iD3/4g8PbPvjgAxQUFOC6665Dbm4ujh07ZnV7S0sL1Gq11c/UajVaW1ud3q5Sqcy38xEX5761vb+JcUy+EirnGirnCYTOudJ5+sb0hGiHP79LG4XfjBuA/Wcvo6apHQlRKowd2AsAsOmXavza0A5Hcy4SAL01KkwZlWwOIpbMGo5HPj0EAHb3kQB4a/ZIJF4ZR0dlI6dxd0il0God/67Y53NVhkgC4NXbrsH04X04PZ834uKikNqXW+CT2jfG6XmJnd8Dn5ycHJw+fdrhbZs2bcKpU6fw+eefO7w9PDwcbW1tVj9ra2tDREQEgO4gqL293er29vZ28+181NU1gfGuIKdgJJLuN6SYxuQroXKuoXKeQOicK51nYKVrlEjXdK8G1F9uBgAszB2EZ5zsNGKu3M4eCwDZvSPxlpMZpKcmpSK7dyR0uiYAQJjJxGlcYSaT+T62nD2fq+f1BcvXdFCUgtPW9EFRCp+OyRPsebgjqu3sGzduxNmzZzFhwgQAQEdHB4xGI7Kzs7Fp0yakp6fj3Llz6OrqglzePfSysjKkp6cDADIyMvDLL79YPWZZWRmGDx/OeywMA1H9UQPiHJOvhMq5hsp5AqFzrnSe4jHJTX2iSelau3OYlK7Fjey2/hYDUvvGYFCUAlKJxOpYPrVrXP2eLJ+vpqkD9W2diFUrkGBRLdtfv2eGAaQSblvTbX8fPYmoAp9PPvnE6t/vv/8+9u/fjzVr1gAAEhISEBsbi3fffRdPPPEEzp49izVr1mDhwoUAgPz8fKxevRpbt27F1KlTsX37duzfvx+LFy/2+7kQQggJPE8qGbMJuhIJoNVGQaezn90SsnaNUAUHhcK3oGVPI6rAxx25XI5Vq1ZhyZIluO666xAeHo65c+fijjvuAACkpqbigw8+wDvvvIPFixcjKSkJ77//PgYOHBjgkRNCCAkUXwUWwRwgBKL1hb9IGKanTlb5lqMIP1BcfesINqFyrqFynkDonCudZ/Dheq6eVG4Wk2B5TdnzcKdHzfgQQgghYiO2pSrimigLGBJCCCGE+AIFPoQQQggJGRT4EEIIISRkUOBDCCGEkJBBgQ8hhBBCQgYFPoQQQggJGRT4EEIIISRkUOBDCCGEkJBBBQydkIio6CY7FjGNyVdC5VxD5TyB0DlXOs/gEyrnGiznyXX81LKCEEIIISGDlroIIYQQEjIo8CGEEEJIyKDAhxBCCCEhgwIfQgghhIQMCnwIIYQQEjIo8CGEEEJIyKDAhxBCCCEhgwIfQgghhIQMCnwIIYQQEjIo8AmwrVu3YtiwYcjMzDT/9/TTTwMAjh49ijvvvBOZmZnIy8vDl19+aXXfDRs2YMqUKRg1ahTuuOMOHD58OBCn4Nbly5cxZcoUFBUVmX/mzbkZjUa89dZbmDBhAjIzM/HII4+gpqbGb+fjjKPzfPnllzF8+HCr13fdunXm23vaeZ46dQr33nsvxo4di+uuuw7PPPMMLl++DCC4XlNX5xlsr+nevXtx5513YvTo0bjuuuvw2muvob29HUBwvaauzjPYXlN2XHPnzsVzzz1n/lkwvZ5eYUhAvfnmm8xzzz1n93O9Xs+MHTuW+fTTT5nOzk5mz549TGZmJnP06FGGYRhm3759TGZmJnPw4EHGYDAwq1evZnJycpjW1lZ/n4JLBw8eZCZPnsxkZGQw+/btYxjG+3N7//33mZkzZzIXL15kmpqamCeeeIJ58MEHA3aODOP4PBmGYW6//Xbmq6++cnifnnaebW1tzHXXXcf87W9/Yzo6OpjLly8zDz74IPPwww8H1Wvq6jwZJrhe07q6OmbEiBHMf/7zH8ZoNDLV1dXMjBkzmL/97W9B9Zq6Ok+GCa7XlPXXv/6VGTJkCPPss88yDBOcn7ueosAnwObMmcN8+umndj//4osvmKlTp1r97KWXXmKeeeYZhmEY5qmnnmJeeOEFq9unT5/OrF+/3neD5emrr75iJk6cyHzzzTdWAYG353bjjTcymzZtMt9WW1vLDB48mLlw4YIvT8cpZ+fZ0dHBXHPNNUxJSYnD+/W08ywvL2fuv/9+pqury/yzHTt2MKNHjw6q19TVeQbba8owDNPU1MQwDMOYTCbm9OnTzJQpU5g1a9YE1WvKMM7PMxhf0z179jC33HILs2DBAnPgE2yvpzdoqSuATCYTfvnlF+zevRuTJk3CjTfeiBdffBENDQ0oLS1FRkaG1fFpaWk4deoUAKCsrMzl7WJw/fXX4/vvv8ctt9xi9XNvzq2pqQm//vqr1e1arRYajQanT5/20Zm45uw8T506ha6uLrz33nuYMGECpk2bhhUrVsBkMgHoeec5aNAg/OMf/4BMJjP/bNu2bbjmmmuC6jV1dZ7B9poCQGRkJAAgNzcXM2fORHx8PO64446gek0B5+cZbK9pXV0dFi9ejHfffRdqtdr882B7Pb1BgU8AXb58GcOGDcO0adOwdetWfP755zh37hyefvpptLS0WL1pAUClUqG1tRUA3N4uBvHx8ZDL5XY/9+bcWlpaAADh4eF2t7O3+Zuz82xqasLYsWMxd+5cFBYW4u2338aaNWuwatUqAD3vPC0xDIPly5dj165dWLx4cdC9pizb8wzm13T79u3473//C6lUigULFgTta2p7nsH0mppMJjz99NO49957MWTIEKvbgvX19AQFPgGk1Wqxdu1azJ49G2q1Gn379sXTTz+N//73v2AYxpx4x2pvb0dERAQAQK1Wu7xdzNyN3dXt7B9mW1ub0/uLxXXXXYd//etfGDt2LBQKBUaOHIk//vGP2Lp1K4Cee57Nzc1YsGABNm/ejE8//RSDBw8OytfU0XkG62sKdF/EEhMT8fTTT+OHH34IytcUsD/P4cOHB81r+vHHH0OpVGLu3Ll2twXr6+kJCnwC6NSpU3jnnXfAMIz5ZwaDAVKpFCNHjkRpaanV8WVlZUhPTwcApKenu7xdzDIyMjw+N41Gg8TERJSVlZlvq62thV6vt5umDbQdO3bg888/t/qZwWCASqUC0DPP88KFC/jNb36D5uZmrF+/HoMHDwYQfK+ps/MMttf00KFDmD59OgwGg/lnBoMBCoUCaWlpQfOaujrPn376KWhe040bN2L//v3Izs5GdnY2tmzZgi1btiA7Ozvo/ka9EtgUo9B26dIlZtSoUcyKFSuYzs5OpqqqirnrrruY559/nrl8+TKTnZ3NrF69mjEYDMzevXuZzMxMZu/evQzDMOaM/L1795oz8MeMGcPU19cH9qScsEz69fbcli9fzsyYMYO5cOGCeXfBPffcE6hTs2J5ntu3b2dGjhzJ7NmzhzGZTMyhQ4eYnJwc5uuvv2YYpuedp16vZyZOnMg899xzjNFotLotmF5TV+cZbK9pc3Mzk5uby7zxxhtMR0cHU1lZycyePZt5+eWXg+o1dXWewfaaWnr22WfNyc3B9Hp6iwKfACsqKmJ++9vfMpmZmcy4ceOY1157jWlvb2cYhmGOHTtmvu2mm25i/vOf/1jd9+uvv2amTZvGjBo1ipk9ezZz5MiRQJwCJ7bbvL05N4PBwLz99tvMDTfcwIwePZp55JFHGJ1O57dzccX2PD/77DNm6tSpzLXXXsvcdNNNdjv4etJ5rlq1isnIyGCuvfZaZtSoUVb/MUzwvKbuzjOYXlOGYZjS0lLm3nvvZbKzs5lJkyYxy5YtYzo6OhiGCZ7XlGFcn2ewvaYsy8CHYYLr9fSGhGEs1lkIIYQQQoIY5fgQQgghJGRQ4EMIIYSQkEGBDyGEEEJCBgU+hBBCCAkZFPgQQgghJGRQ4EMIIYSQkEGBDyGEEEJCBgU+hJAe7dy5c4EeAiGkB6HAhxDSYzAMg6eeegqjRo1CXl4eTpw4gRkzZnC+f15eHr766isAwAMPPICPPvrI7X1eeuklvPTSSx6PmRAiLvJAD4AQQriqqanBli1b8NVXX+Gaa65BUVEROjs7PXqsf/zjH5yOW7JkiUePTwgRJwp8CCEB8f7772P9+vVoa2tDSkoKHn30Udx0003YuXMnli1bhqqqKowePRopKSno6OjAH/7wB/z+978HAMyZMwdTpkzBtm3bAACZmZlYtWoVMjMzOT//3LlzMXbsWNx2222YOnUqtm7dikGDBgEAysvLkZ+fj127dmHZsmUAgDfffBPvv/8+SktLoVQqsXv3boSHh+O2227DU089BQBob2/H0qVL8e2330KtVuP222/Hpk2bsHTpUuTk5Aj56yOEeIiWugghfrdv3z6sW7cOX375JYqKinDnnXdi8eLFOH36NB5//HE8/PDDOHjwIO666y6sX78eADBs2DBs2bIFALBlyxa8/fbbWLlyJQDg8OHDvIIeS/369UNOTg42btxo/tlXX32FG264AQkJCXbHb9++Hddffz2Kiorw2muvYeXKlThy5AgA4I033sDx48exceNGbN26FRcvXkRVVZVH4yKE+AYFPoQQvwsLC0NDQwO++OILnDhxAnfeeSf27t2Lbdu2Yfjw4cjPz4dcLsf06dMxceJEn4/nzjvvxKZNm8AwDIxGIzZt2oTZs2c7PHbAgAGYNWsWZDIZcnNzER8fj3PnzqGzsxObNm3CwoUL0adPH0REROCll16CTCbz+fgJIdxR4EMI8bvMzEy8//77OHz4MObMmYPrrrsOH374Ierq6tC3b1+rYwcOHOjz8UydOhWtra0oKirCjz/+CIZhnAZc8fHxVv9WKBQwmUzQ6/Voa2tDUlKS+bbIyEjExsb6cuiEEJ4ox4cQ4ncXL15EXFwcPvnkExgMBuzduxf/8z//g4cffhgnTpywOvbXX3+FXO7bjyqlUon8/Hxs2bIFbW1tmDVrFu/njIuLg0qlwsWLF825Qq2traivr/fFkAkhHqIZH0KI3x0/fhwPPPAATp06BaVSibi4OADA2LFjcebMGaxbtw5dXV3Ys2ePOYHZkbCwMABAU1OT12O66667sGPHDhQUFDhd5nJFKpVi9uzZeP/991FdXY22tjYsXboURqPR67ERQoRDgQ8hxO+mTZuG++67D4888ghGjRqFxx9/HM8//zzGjh2L1atXY8OGDRg3bhxWrlyJMWPGOH2cjIwMZGVl4YYbbkBhYaFXY0pPT8eAAQNwzTXXYMCAAR49xlNPPYVBgwbhlltuwbRp09C7d29IpVIoFAqvxkYIEY6EYRgm0IMghBBnnnvuOQDd28nF7sCBAxg8eDCio6MBAM3NzcjKysK2bds8DqYIIcKiGR9CCBHIqlWr8Prrr6O9vR0dHR147733MHDgQAp6CBERSm4mhASF+fPnY8+ePU5vf/XVV5Gfn+/TMbzyyit49dVXkZubC6PRiKysLKxYscKnz0kI4YeWugghhBASMmipixBCCCEhgwIfQgghhIQMCnwIIYQQEjIo8CGEEEJIyKDAhxBCCCEhgwIfQgghhIQMCnwIIYQQEjIo8CGEEEJIyKDAhxBCCCEh4/8DSmHTPGFR7IoAAAAASUVORK5CYII=", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# plot the residuals \n", + "fig, ax = plt.subplots()\n", + "\n", + "ax.scatter(modeling_df.sqft_living, model.resid)\n", + "ax.axhline(y=0, color=\"black\")\n", + "ax.set_xlabel(\"sqft_living\")\n", + "ax.set_ylabel(\"residuals\");" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We create a qq plot to examine the residuals of our model." + ] + }, + { + "cell_type": "code", + "execution_count": 86, + "metadata": {}, + "outputs": [], + "source": [ + "from scipy import stats\n", + "\n", + "def model_qq_plot(model):\n", + " sm.graphics.qqplot(model.resid, dist=stats.norm, line='45', fit=True)\n", + " plt.show();\n" + ] + }, + { + "cell_type": "code", + "execution_count": 87, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "model_qq_plot(model)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "As the points roughly follow the line, it is signalling the distribution is normal." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Multiple Linear Regression" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Next we test if adding all columns will improve the model." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "A function that returns all independent columns in formula form and another that creates a model using all independent variables." ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 308, "metadata": {}, "outputs": [], "source": [ - "# Your code here - remember to use markdown cells for comments as well!" + "def all_columns():\n", + " return \"+\".join(modeling_df.columns.difference([\"price\"]))\n", + "all_columns()\n", + "\n", + "def model_all_independent():\n", + " multi_formula = \"price ~ \"+all_columns()\n", + " return ols(multi_formula, modeling_df).fit()" + ] + }, + { + "cell_type": "code", + "execution_count": 309, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " OLS Regression Results \n", + "==============================================================================\n", + "Dep. Variable: price R-squared: 0.518\n", + "Model: OLS Adj. R-squared: 0.510\n", + "Method: Least Squares F-statistic: 70.47\n", + "Date: Thu, 02 May 2024 Prob (F-statistic): 0.00\n", + "Time: 02:43:07 Log-Likelihood: -1.3708e+05\n", + "No. Observations: 21534 AIC: 2.748e+05\n", + "Df Residuals: 21210 BIC: 2.774e+05\n", + "Df Model: 323 \n", + "Covariance Type: nonrobust \n", + "===========================================================================================\n", + " coef std err t P>|t| [0.025 0.975]\n", + "-------------------------------------------------------------------------------------------\n", + "Intercept 4556.8611 91.218 49.956 0.000 4378.067 4735.655\n", + "grade[T.11] -9.3225 8.574 -1.087 0.277 -26.128 7.483\n", + "grade[T.12] -70.8463 17.170 -4.126 0.000 -104.501 -37.192\n", + "grade[T.13] -108.1645 47.546 -2.275 0.023 -201.357 -14.972\n", + "grade[T.3] -311.3480 142.047 -2.192 0.028 -589.771 -32.925\n", + "grade[T.4] -368.6820 28.146 -13.099 0.000 -423.850 -313.514\n", + "grade[T.5] -356.5435 11.198 -31.840 0.000 -378.493 -334.594\n", + "grade[T.6] -310.7058 6.953 -44.686 0.000 -324.334 -297.077\n", + "grade[T.7] -225.6585 5.655 -39.906 0.000 -236.742 -214.575\n", + "grade[T.8] -132.2447 5.124 -25.809 0.000 -142.288 -122.201\n", + "grade[T.9] -35.4235 5.188 -6.828 0.000 -45.593 -25.254\n", + "sqft_basement[T.0.0] -1.0952 6.839 -0.160 0.873 -14.500 12.310\n", + "sqft_basement[T.10.0] -118.6282 100.574 -1.180 0.238 -315.760 78.504\n", + "sqft_basement[T.100.0] 54.0306 22.928 2.357 0.018 9.090 98.972\n", + "sqft_basement[T.1000.0] 5.5236 13.656 0.404 0.686 -21.244 32.291\n", + "sqft_basement[T.1008.0] -145.1546 142.100 -1.021 0.307 -423.682 133.372\n", + "sqft_basement[T.1010.0] -4.8311 19.162 -0.252 0.801 -42.390 32.728\n", + "sqft_basement[T.1020.0] 29.2003 21.087 1.385 0.166 -12.133 70.533\n", + "sqft_basement[T.1024.0] -40.8944 142.128 -0.288 0.774 -319.475 237.687\n", + "sqft_basement[T.1030.0] -16.2143 22.568 -0.718 0.472 -60.450 28.021\n", + "sqft_basement[T.1040.0] -1.9852 20.368 -0.097 0.922 -41.909 37.938\n", + "sqft_basement[T.1050.0] -5.3511 23.531 -0.227 0.820 -51.473 40.771\n", + "sqft_basement[T.1060.0] 3.6221 19.925 0.182 0.856 -35.433 42.677\n", + "sqft_basement[T.1070.0] 29.0525 21.078 1.378 0.168 -12.262 70.367\n", + "sqft_basement[T.1080.0] 4.7594 26.078 0.183 0.855 -46.356 55.875\n", + "sqft_basement[T.1090.0] -40.7648 26.063 -1.564 0.118 -91.849 10.320\n", + "sqft_basement[T.110.0] 1.3419 34.131 0.039 0.969 -65.557 68.240\n", + "sqft_basement[T.1100.0] 22.5876 17.522 1.289 0.197 -11.756 56.932\n", + "sqft_basement[T.1110.0] 34.0421 25.034 1.360 0.174 -15.026 83.110\n", + "sqft_basement[T.1120.0] -7.9712 23.025 -0.346 0.729 -53.102 37.160\n", + "sqft_basement[T.1130.0] 4.1381 26.873 0.154 0.878 -48.536 56.812\n", + "sqft_basement[T.1135.0] -182.0744 142.064 -1.282 0.200 -460.531 96.382\n", + "sqft_basement[T.1140.0] -45.0422 28.728 -1.568 0.117 -101.351 11.267\n", + "sqft_basement[T.1150.0] 25.4643 28.199 0.903 0.367 -29.807 80.736\n", + "sqft_basement[T.1160.0] -111.0285 28.734 -3.864 0.000 -167.349 -54.708\n", + "sqft_basement[T.1170.0] -10.9192 26.877 -0.406 0.685 -63.600 41.761\n", + "sqft_basement[T.1180.0] -23.3639 27.768 -0.841 0.400 -77.792 31.064\n", + "sqft_basement[T.1190.0] -23.2185 29.862 -0.778 0.437 -81.751 35.314\n", + "sqft_basement[T.120.0] 35.1522 20.628 1.704 0.088 -5.280 75.585\n", + "sqft_basement[T.1200.0] -13.2539 18.543 -0.715 0.475 -49.599 23.091\n", + "sqft_basement[T.1210.0] -67.9431 34.169 -1.988 0.047 -134.916 -0.970\n", + "sqft_basement[T.1220.0] 5.2166 26.458 0.197 0.844 -46.643 57.076\n", + "sqft_basement[T.1230.0] 12.2156 26.847 0.455 0.649 -40.406 64.837\n", + "sqft_basement[T.1240.0] 27.6897 32.467 0.853 0.394 -35.949 91.328\n", + "sqft_basement[T.1245.0] -61.3090 142.062 -0.432 0.666 -339.761 217.143\n", + "sqft_basement[T.1248.0] 55.4228 142.072 0.390 0.696 -223.048 333.894\n", + "sqft_basement[T.1250.0] -38.7097 21.353 -1.813 0.070 -80.562 3.143\n", + "sqft_basement[T.1260.0] -35.3169 31.793 -1.111 0.267 -97.634 27.000\n", + "sqft_basement[T.1270.0] -47.6132 26.910 -1.769 0.077 -100.359 5.132\n", + "sqft_basement[T.1275.0] 376.3214 142.105 2.648 0.008 97.784 654.858\n", + "sqft_basement[T.1280.0] 11.0876 31.057 0.357 0.721 -49.786 71.961\n", + "sqft_basement[T.1281.0] -337.7820 142.100 -2.377 0.017 -616.309 -59.255\n", + "sqft_basement[T.1284.0] 110.2182 142.238 0.775 0.438 -168.579 389.015\n", + "sqft_basement[T.1290.0] -25.3371 33.345 -0.760 0.447 -90.696 40.022\n", + "sqft_basement[T.130.0] 43.4859 29.188 1.490 0.136 -13.725 100.697\n", + "sqft_basement[T.1300.0] 20.5723 23.275 0.884 0.377 -25.049 66.194\n", + "sqft_basement[T.1310.0] -60.9261 39.988 -1.524 0.128 -139.305 17.453\n", + "sqft_basement[T.1320.0] -82.6295 31.070 -2.660 0.008 -143.528 -21.731\n", + "sqft_basement[T.1330.0] -4.8148 32.503 -0.148 0.882 -68.524 58.894\n", + "sqft_basement[T.1340.0] -29.1984 33.388 -0.875 0.382 -94.641 36.244\n", + "sqft_basement[T.1350.0] -49.9322 37.391 -1.335 0.182 -123.222 23.358\n", + "sqft_basement[T.1360.0] -16.1567 33.356 -0.484 0.628 -81.538 49.224\n", + "sqft_basement[T.1370.0] -49.0166 30.500 -1.607 0.108 -108.799 10.766\n", + "sqft_basement[T.1380.0] -66.4757 31.789 -2.091 0.037 -128.784 -4.168\n", + "sqft_basement[T.1390.0] 36.3204 35.150 1.033 0.301 -32.576 105.216\n", + "sqft_basement[T.140.0] 52.5007 20.629 2.545 0.011 12.065 92.936\n", + "sqft_basement[T.1400.0] 66.4879 24.108 2.758 0.006 19.235 113.741\n", + "sqft_basement[T.1410.0] 24.0433 43.439 0.553 0.580 -61.101 109.187\n", + "sqft_basement[T.1420.0] 41.6786 34.202 1.219 0.223 -25.360 108.717\n", + "sqft_basement[T.143.0] 130.6261 142.032 0.920 0.358 -147.768 409.020\n", + "sqft_basement[T.1430.0] 54.1768 40.131 1.350 0.177 -24.483 132.837\n", + "sqft_basement[T.1440.0] 45.0015 43.381 1.037 0.300 -40.028 130.031\n", + "sqft_basement[T.145.0] 47.3116 63.837 0.741 0.459 -77.814 172.437\n", + "sqft_basement[T.1450.0] 23.8517 34.268 0.696 0.486 -43.315 91.019\n", + "sqft_basement[T.1460.0] 119.2898 38.568 3.093 0.002 43.694 194.886\n", + "sqft_basement[T.1470.0] -196.1417 63.902 -3.069 0.002 -321.395 -70.888\n", + "sqft_basement[T.1480.0] 27.9043 47.837 0.583 0.560 -65.860 121.669\n", + "sqft_basement[T.1481.0] 39.0470 142.114 0.275 0.784 -239.508 317.602\n", + "sqft_basement[T.1490.0] -47.6760 47.879 -0.996 0.319 -141.522 46.170\n", + "sqft_basement[T.150.0] 16.8237 22.424 0.750 0.453 -27.130 60.777\n", + "sqft_basement[T.1500.0] -25.7490 27.391 -0.940 0.347 -79.438 27.940\n", + "sqft_basement[T.1510.0] 58.9471 43.419 1.358 0.175 -26.158 144.052\n", + "sqft_basement[T.1520.0] -35.5303 54.132 -0.656 0.512 -141.634 70.573\n", + "sqft_basement[T.1525.0] -151.9416 142.066 -1.070 0.285 -430.402 126.519\n", + "sqft_basement[T.1530.0] 73.3914 58.366 1.257 0.209 -41.010 187.793\n", + "sqft_basement[T.1540.0] -4.8143 43.505 -0.111 0.912 -90.088 80.459\n", + "sqft_basement[T.1548.0] 65.0450 142.117 0.458 0.647 -213.516 343.606\n", + "sqft_basement[T.1550.0] -126.5743 63.893 -1.981 0.048 -251.809 -1.340\n", + "sqft_basement[T.1560.0] -146.5136 54.181 -2.704 0.007 -252.713 -40.314\n", + "sqft_basement[T.1570.0] 25.5016 50.713 0.503 0.615 -73.900 124.903\n", + "sqft_basement[T.1580.0] 7.0502 41.691 0.169 0.866 -74.667 88.768\n", + "sqft_basement[T.1590.0] -74.5863 40.008 -1.864 0.062 -153.005 3.832\n", + "sqft_basement[T.160.0] 24.2421 27.195 0.891 0.373 -29.063 77.547\n", + "sqft_basement[T.1600.0] -50.1041 35.186 -1.424 0.154 -119.072 18.864\n", + "sqft_basement[T.1610.0] -80.0352 71.401 -1.121 0.262 -219.987 59.917\n", + "sqft_basement[T.1620.0] -72.8020 54.181 -1.344 0.179 -179.002 33.397\n", + "sqft_basement[T.1630.0] -107.4601 71.380 -1.505 0.132 -247.371 32.450\n", + "sqft_basement[T.1640.0] 59.4576 64.255 0.925 0.355 -66.486 185.402\n", + "sqft_basement[T.1650.0] 3.5031 47.921 0.073 0.942 -90.426 97.432\n", + "sqft_basement[T.1660.0] 59.9727 54.122 1.108 0.268 -46.110 166.056\n", + "sqft_basement[T.1670.0] -28.1349 63.878 -0.440 0.660 -153.340 97.071\n", + "sqft_basement[T.1680.0] -44.4094 58.437 -0.760 0.447 -158.951 70.132\n", + "sqft_basement[T.1690.0] -40.5300 50.753 -0.799 0.425 -140.009 58.949\n", + "sqft_basement[T.170.0] 43.7006 27.646 1.581 0.114 -10.487 97.888\n", + "sqft_basement[T.1700.0] 18.5867 54.122 0.343 0.731 -87.496 124.669\n", + "sqft_basement[T.1710.0] 54.6206 63.924 0.854 0.393 -70.676 179.917\n", + "sqft_basement[T.172.0] 128.7633 142.034 0.907 0.365 -149.635 407.162\n", + "sqft_basement[T.1720.0] 59.6116 54.252 1.099 0.272 -46.727 165.950\n", + "sqft_basement[T.1730.0] -4.1005 100.674 -0.041 0.968 -201.429 193.228\n", + "sqft_basement[T.1740.0] 126.0634 82.278 1.532 0.125 -35.207 287.334\n", + "sqft_basement[T.1750.0] -36.9714 82.228 -0.450 0.653 -198.144 124.201\n", + "sqft_basement[T.176.0] -107.0199 142.060 -0.753 0.451 -385.468 171.428\n", + "sqft_basement[T.1760.0] -147.5997 50.846 -2.903 0.004 -247.262 -47.937\n", + "sqft_basement[T.1770.0] -50.0119 142.141 -0.352 0.725 -328.620 228.596\n", + "sqft_basement[T.1780.0] -115.3663 50.685 -2.276 0.023 -214.714 -16.019\n", + "sqft_basement[T.1790.0] -40.0153 58.398 -0.685 0.493 -154.481 74.450\n", + "sqft_basement[T.1798.0] -68.5047 142.106 -0.482 0.630 -347.044 210.035\n", + "sqft_basement[T.180.0] 86.3894 23.991 3.601 0.000 39.365 133.413\n", + "sqft_basement[T.1800.0] -0.9385 43.454 -0.022 0.983 -86.111 84.234\n", + "sqft_basement[T.1810.0] -104.2988 100.584 -1.037 0.300 -301.452 92.854\n", + "sqft_basement[T.1816.0] -6.5198 143.161 -0.046 0.964 -287.126 274.086\n", + "sqft_basement[T.1820.0] -134.2476 100.594 -1.335 0.182 -331.420 62.925\n", + "sqft_basement[T.1830.0] -38.1477 63.860 -0.597 0.550 -163.319 87.023\n", + "sqft_basement[T.1840.0] -27.1976 142.102 -0.191 0.848 -305.728 251.333\n", + "sqft_basement[T.1850.0] -170.3039 58.548 -2.909 0.004 -285.063 -55.545\n", + "sqft_basement[T.1852.0] -251.4916 142.237 -1.768 0.077 -530.287 27.304\n", + "sqft_basement[T.1860.0] 30.6464 83.755 0.366 0.714 -133.520 194.813\n", + "sqft_basement[T.1870.0] -120.0403 58.404 -2.055 0.040 -234.517 -5.564\n", + "sqft_basement[T.1880.0] -226.6599 142.236 -1.594 0.111 -505.453 52.134\n", + "sqft_basement[T.1890.0] 12.9115 100.917 0.128 0.898 -184.893 210.716\n", + "sqft_basement[T.190.0] 25.8236 25.633 1.007 0.314 -24.419 76.066\n", + "sqft_basement[T.1900.0] -0.1395 71.329 -0.002 0.998 -139.949 139.670\n", + "sqft_basement[T.1910.0] -80.2410 82.270 -0.975 0.329 -241.495 81.013\n", + "sqft_basement[T.1913.0] -147.9231 142.120 -1.041 0.298 -426.488 130.642\n", + "sqft_basement[T.1920.0] -210.3262 142.179 -1.479 0.139 -489.007 68.355\n", + "sqft_basement[T.1930.0] 76.5646 142.130 0.539 0.590 -202.021 355.150\n", + "sqft_basement[T.1940.0] -31.5902 64.063 -0.493 0.622 -157.158 93.978\n", + "sqft_basement[T.1950.0] -10.0209 71.437 -0.140 0.888 -150.042 130.000\n", + "sqft_basement[T.1960.0] -63.9273 142.254 -0.449 0.653 -342.757 214.902\n", + "sqft_basement[T.1990.0] 80.3039 142.247 0.565 0.572 -198.512 359.120\n", + "sqft_basement[T.20.0] 24.3534 142.062 0.171 0.864 -254.099 302.805\n", + "sqft_basement[T.200.0] 46.6274 15.413 3.025 0.002 16.416 76.838\n", + "sqft_basement[T.2000.0] 103.4934 100.705 1.028 0.304 -93.895 300.882\n", + "sqft_basement[T.2010.0] 93.5608 100.631 0.930 0.353 -103.684 290.806\n", + "sqft_basement[T.2020.0] -196.6478 63.978 -3.074 0.002 -322.051 -71.245\n", + "sqft_basement[T.2030.0] -205.8635 100.595 -2.046 0.041 -403.037 -8.690\n", + "sqft_basement[T.2040.0] 145.4551 100.707 1.444 0.149 -51.939 342.849\n", + "sqft_basement[T.2050.0] -229.8486 142.255 -1.616 0.106 -508.678 48.981\n", + "sqft_basement[T.2060.0] -92.7652 63.895 -1.452 0.147 -218.003 32.473\n", + "sqft_basement[T.207.0] 79.4642 142.019 0.560 0.576 -198.905 357.833\n", + "sqft_basement[T.2070.0] 10.7868 82.240 0.131 0.896 -150.409 171.983\n", + "sqft_basement[T.2080.0] 22.1989 100.603 0.221 0.825 -174.991 219.389\n", + "sqft_basement[T.2090.0] 44.9722 100.641 0.447 0.655 -152.291 242.236\n", + "sqft_basement[T.210.0] 17.0251 28.663 0.594 0.553 -39.156 73.207\n", + "sqft_basement[T.2100.0] -193.1056 100.760 -1.916 0.055 -390.604 4.392\n", + "sqft_basement[T.2110.0] -180.4200 100.655 -1.792 0.073 -377.711 16.871\n", + "sqft_basement[T.2120.0] -21.8932 142.217 -0.154 0.878 -300.649 256.862\n", + "sqft_basement[T.2130.0] -9.0360 142.320 -0.063 0.949 -287.995 269.922\n", + "sqft_basement[T.2150.0] 287.6115 82.263 3.496 0.000 126.369 448.854\n", + "sqft_basement[T.2160.0] 114.9934 82.298 1.397 0.162 -46.317 276.304\n", + "sqft_basement[T.2170.0] -94.5632 82.363 -1.148 0.251 -256.002 66.875\n", + "sqft_basement[T.2190.0] 238.3882 142.133 1.677 0.094 -40.203 516.979\n", + "sqft_basement[T.2196.0] -30.0403 142.130 -0.211 0.833 -308.626 248.545\n", + "sqft_basement[T.220.0] 27.2540 22.670 1.202 0.229 -17.182 71.690\n", + "sqft_basement[T.2200.0] 93.6441 100.995 0.927 0.354 -104.313 291.601\n", + "sqft_basement[T.2220.0] 67.3267 82.488 0.816 0.414 -94.356 229.009\n", + "sqft_basement[T.2240.0] 45.3178 143.119 0.317 0.752 -235.207 325.842\n", + "sqft_basement[T.225.0] 196.5159 142.024 1.384 0.166 -81.862 474.893\n", + "sqft_basement[T.2250.0] 412.7765 142.125 2.904 0.004 134.201 691.352\n", + "sqft_basement[T.230.0] 32.0814 43.327 0.740 0.459 -52.844 117.007\n", + "sqft_basement[T.2300.0] 155.3742 142.104 1.093 0.274 -123.160 433.908\n", + "sqft_basement[T.2310.0] 46.9855 142.350 0.330 0.741 -232.031 326.002\n", + "sqft_basement[T.2330.0] 45.1976 101.966 0.443 0.658 -154.663 245.058\n", + "sqft_basement[T.235.0] -35.0382 100.555 -0.348 0.728 -232.134 162.057\n", + "sqft_basement[T.2350.0] -72.6321 142.125 -0.511 0.609 -351.207 205.943\n", + "sqft_basement[T.2360.0] -135.2886 142.234 -0.951 0.342 -414.077 143.500\n", + "sqft_basement[T.240.0] 39.0099 17.824 2.189 0.029 4.074 73.946\n", + "sqft_basement[T.2400.0] 91.7244 142.213 0.645 0.519 -187.024 370.473\n", + "sqft_basement[T.243.0] 91.9403 142.045 0.647 0.517 -186.478 370.358\n", + "sqft_basement[T.248.0] -53.6612 142.021 -0.378 0.706 -332.034 224.711\n", + "sqft_basement[T.2490.0] -132.2543 142.330 -0.929 0.353 -411.232 146.723\n", + "sqft_basement[T.250.0] 36.8194 20.616 1.786 0.074 -3.589 77.228\n", + "sqft_basement[T.2500.0] 62.5815 143.147 0.437 0.662 -217.997 343.160\n", + "sqft_basement[T.2550.0] -59.5786 103.424 -0.576 0.565 -262.297 143.140\n", + "sqft_basement[T.2570.0] 11.3598 142.273 0.080 0.936 -267.507 290.226\n", + "sqft_basement[T.2580.0] 33.1542 143.141 0.232 0.817 -247.414 313.722\n", + "sqft_basement[T.260.0] 29.1433 22.900 1.273 0.203 -15.743 74.029\n", + "sqft_basement[T.2600.0] 517.0270 142.324 3.633 0.000 238.060 795.994\n", + "sqft_basement[T.2610.0] -174.9837 142.165 -1.231 0.218 -453.638 103.670\n", + "sqft_basement[T.265.0] 31.1394 82.218 0.379 0.705 -130.013 192.292\n", + "sqft_basement[T.266.0] -25.7475 142.017 -0.181 0.856 -304.113 252.618\n", + "sqft_basement[T.270.0] -9.7889 22.698 -0.431 0.666 -54.278 34.700\n", + "sqft_basement[T.2720.0] 254.6828 142.130 1.792 0.073 -23.902 533.267\n", + "sqft_basement[T.2730.0] 53.4192 149.677 0.357 0.721 -239.959 346.798\n", + "sqft_basement[T.274.0] -46.9447 142.043 -0.330 0.741 -325.360 231.471\n", + "sqft_basement[T.276.0] 36.5025 142.042 0.257 0.797 -241.910 314.915\n", + "sqft_basement[T.280.0] 28.9333 19.236 1.504 0.133 -8.770 66.637\n", + "sqft_basement[T.2810.0] 59.2155 142.973 0.414 0.679 -221.023 339.454\n", + "sqft_basement[T.283.0] 52.7916 142.095 0.372 0.710 -225.726 331.309\n", + "sqft_basement[T.2850.0] -73.3197 142.248 -0.515 0.606 -352.137 205.498\n", + "sqft_basement[T.290.0] -2.1455 18.370 -0.117 0.907 -38.151 33.860\n", + "sqft_basement[T.295.0] 31.0690 142.114 0.219 0.827 -247.486 309.624\n", + "sqft_basement[T.300.0] 29.3597 13.755 2.135 0.033 2.399 56.320\n", + "sqft_basement[T.3000.0] 13.6840 149.711 0.091 0.927 -279.761 307.129\n", + "sqft_basement[T.310.0] -2.4762 19.812 -0.125 0.901 -41.309 36.357\n", + "sqft_basement[T.320.0] 55.4037 23.426 2.365 0.018 9.487 101.321\n", + "sqft_basement[T.3260.0] -28.0316 142.307 -0.197 0.844 -306.964 250.901\n", + "sqft_basement[T.330.0] -7.4864 21.763 -0.344 0.731 -50.143 35.171\n", + "sqft_basement[T.340.0] 10.3779 19.225 0.540 0.589 -27.305 48.061\n", + "sqft_basement[T.3480.0] -162.2113 149.698 -1.084 0.279 -455.631 131.209\n", + "sqft_basement[T.350.0] 3.1173 18.589 0.168 0.867 -33.319 39.554\n", + "sqft_basement[T.3500.0] 29.7954 143.202 0.208 0.835 -250.892 310.483\n", + "sqft_basement[T.360.0] -11.6775 18.131 -0.644 0.520 -47.216 23.861\n", + "sqft_basement[T.370.0] 31.6936 23.149 1.369 0.171 -13.681 77.068\n", + "sqft_basement[T.374.0] -98.3864 142.063 -0.693 0.489 -376.842 180.069\n", + "sqft_basement[T.380.0] 17.9135 17.922 1.000 0.318 -17.214 53.041\n", + "sqft_basement[T.390.0] 42.0538 21.974 1.914 0.056 -1.017 85.124\n", + "sqft_basement[T.40.0] -52.7710 71.297 -0.740 0.459 -192.519 86.977\n", + "sqft_basement[T.400.0] -0.5337 12.431 -0.043 0.966 -24.899 23.831\n", + "sqft_basement[T.410.0] 4.6765 29.727 0.157 0.875 -53.590 62.943\n", + "sqft_basement[T.4130.0] -8.0521 143.050 -0.056 0.955 -288.442 272.338\n", + "sqft_basement[T.414.0] 52.7790 100.552 0.525 0.600 -144.311 249.868\n", + "sqft_basement[T.415.0] 133.1458 142.034 0.937 0.349 -145.252 411.543\n", + "sqft_basement[T.417.0] -410.5828 142.089 -2.890 0.004 -689.087 -132.079\n", + "sqft_basement[T.420.0] 23.9271 17.145 1.396 0.163 -9.679 57.533\n", + "sqft_basement[T.430.0] 56.5876 18.974 2.982 0.003 19.398 93.777\n", + "sqft_basement[T.435.0] 6.8322 100.571 0.068 0.946 -190.294 203.958\n", + "sqft_basement[T.440.0] 10.3948 18.366 0.566 0.571 -25.605 46.394\n", + "sqft_basement[T.450.0] 28.2132 15.511 1.819 0.069 -2.190 58.616\n", + "sqft_basement[T.460.0] -11.4295 19.520 -0.586 0.558 -49.690 26.831\n", + "sqft_basement[T.470.0] 8.3618 20.982 0.399 0.690 -32.765 49.488\n", + "sqft_basement[T.475.0] 136.5105 142.024 0.961 0.336 -141.867 414.887\n", + "sqft_basement[T.480.0] 13.3432 15.572 0.857 0.392 -17.178 43.865\n", + "sqft_basement[T.4820.0] -67.8855 143.102 -0.474 0.635 -348.375 212.604\n", + "sqft_basement[T.490.0] 33.4988 27.649 1.212 0.226 -20.695 87.693\n", + "sqft_basement[T.50.0] 64.4462 43.344 1.487 0.137 -20.512 149.405\n", + "sqft_basement[T.500.0] 13.8606 11.912 1.164 0.245 -9.488 37.210\n", + "sqft_basement[T.506.0] -111.4229 142.035 -0.784 0.433 -389.822 166.976\n", + "sqft_basement[T.508.0] 219.5180 142.031 1.546 0.122 -58.873 497.909\n", + "sqft_basement[T.510.0] 5.9629 21.566 0.276 0.782 -36.307 48.233\n", + "sqft_basement[T.515.0] 167.6865 100.579 1.667 0.095 -29.456 364.829\n", + "sqft_basement[T.516.0] -51.0725 142.028 -0.360 0.719 -329.457 227.312\n", + "sqft_basement[T.518.0] 69.5462 142.028 0.490 0.624 -208.840 347.932\n", + "sqft_basement[T.520.0] 35.6299 18.481 1.928 0.054 -0.594 71.854\n", + "sqft_basement[T.530.0] -19.7624 15.531 -1.272 0.203 -50.205 10.680\n", + "sqft_basement[T.540.0] 11.6519 20.280 0.575 0.566 -28.099 51.403\n", + "sqft_basement[T.550.0] -14.1005 17.624 -0.800 0.424 -48.646 20.445\n", + "sqft_basement[T.556.0] -232.2094 142.109 -1.634 0.102 -510.754 46.335\n", + "sqft_basement[T.560.0] 15.8113 19.242 0.822 0.411 -21.905 53.527\n", + "sqft_basement[T.570.0] 33.7106 19.248 1.751 0.080 -4.016 71.437\n", + "sqft_basement[T.580.0] 1.6078 16.897 0.095 0.924 -31.512 34.727\n", + "sqft_basement[T.588.0] -194.1628 142.057 -1.367 0.172 -472.604 84.279\n", + "sqft_basement[T.590.0] 20.0979 23.163 0.868 0.386 -25.303 65.499\n", + "sqft_basement[T.60.0] -7.7807 45.390 -0.171 0.864 -96.748 81.187\n", + "sqft_basement[T.600.0] 4.6500 11.806 0.394 0.694 -18.491 27.791\n", + "sqft_basement[T.602.0] -124.5024 142.035 -0.877 0.381 -402.901 153.897\n", + "sqft_basement[T.610.0] 32.9453 22.207 1.484 0.138 -10.582 76.473\n", + "sqft_basement[T.620.0] 15.7428 16.430 0.958 0.338 -16.461 47.947\n", + "sqft_basement[T.630.0] 20.5088 18.737 1.095 0.274 -16.216 57.234\n", + "sqft_basement[T.640.0] -21.4124 19.119 -1.120 0.263 -58.886 16.062\n", + "sqft_basement[T.65.0] 16.6854 142.063 0.117 0.907 -261.768 295.139\n", + "sqft_basement[T.650.0] 15.7391 17.736 0.887 0.375 -19.025 50.503\n", + "sqft_basement[T.652.0] 199.6126 142.016 1.406 0.160 -78.749 477.974\n", + "sqft_basement[T.660.0] 76.1234 21.572 3.529 0.000 33.841 118.406\n", + "sqft_basement[T.666.0] -265.2152 142.044 -1.867 0.062 -543.632 13.202\n", + "sqft_basement[T.670.0] -21.4445 17.541 -1.223 0.222 -55.826 12.938\n", + "sqft_basement[T.680.0] -3.3273 17.949 -0.185 0.853 -38.509 31.854\n", + "sqft_basement[T.690.0] 33.3591 22.222 1.501 0.133 -10.198 76.916\n", + "sqft_basement[T.70.0] -7.3489 58.355 -0.126 0.900 -121.729 107.031\n", + "sqft_basement[T.700.0] -17.9272 11.997 -1.494 0.135 -41.442 5.588\n", + "sqft_basement[T.704.0] -70.5983 142.235 -0.496 0.620 -349.389 208.193\n", + "sqft_basement[T.710.0] 35.7155 24.303 1.470 0.142 -11.920 83.351\n", + "sqft_basement[T.720.0] -19.8003 15.890 -1.246 0.213 -50.946 11.346\n", + "sqft_basement[T.730.0] 28.3061 19.010 1.489 0.137 -8.955 65.567\n", + "sqft_basement[T.740.0] -3.6663 19.024 -0.193 0.847 -40.955 33.622\n", + "sqft_basement[T.750.0] 3.9122 15.526 0.252 0.801 -26.519 34.344\n", + "sqft_basement[T.760.0] 20.2084 19.142 1.056 0.291 -17.311 57.727\n", + "sqft_basement[T.768.0] -298.2700 142.091 -2.099 0.036 -576.779 -19.761\n", + "sqft_basement[T.770.0] 9.2442 18.194 0.508 0.611 -26.418 44.906\n", + "sqft_basement[T.780.0] 10.2736 17.771 0.578 0.563 -24.560 45.107\n", + "sqft_basement[T.784.0] -185.8995 142.037 -1.309 0.191 -464.302 92.503\n", + "sqft_basement[T.790.0] -18.3129 19.567 -0.936 0.349 -56.666 20.040\n", + "sqft_basement[T.792.0] 71.9782 142.069 0.507 0.612 -206.488 350.445\n", + "sqft_basement[T.80.0] 2.3984 32.520 0.074 0.941 -61.343 66.140\n", + "sqft_basement[T.800.0] 12.9939 12.144 1.070 0.285 -10.809 36.797\n", + "sqft_basement[T.810.0] -10.5736 20.503 -0.516 0.606 -50.761 29.614\n", + "sqft_basement[T.820.0] 33.8076 19.569 1.728 0.084 -4.550 72.165\n", + "sqft_basement[T.830.0] -7.0886 20.325 -0.349 0.727 -46.928 32.750\n", + "sqft_basement[T.840.0] -4.1195 17.045 -0.242 0.809 -37.529 29.290\n", + "sqft_basement[T.850.0] 44.0507 18.206 2.420 0.016 8.365 79.736\n", + "sqft_basement[T.860.0] 3.0138 17.501 0.172 0.863 -31.289 37.317\n", + "sqft_basement[T.861.0] -272.7353 142.091 -1.919 0.055 -551.245 5.774\n", + "sqft_basement[T.862.0] -106.5600 142.045 -0.750 0.453 -384.979 171.859\n", + "sqft_basement[T.870.0] 24.7255 21.614 1.144 0.253 -17.639 67.090\n", + "sqft_basement[T.875.0] 124.5565 142.213 0.876 0.381 -154.192 403.305\n", + "sqft_basement[T.880.0] 44.9721 18.543 2.425 0.015 8.626 81.318\n", + "sqft_basement[T.890.0] 7.7268 20.881 0.370 0.711 -33.201 48.654\n", + "sqft_basement[T.90.0] 13.1625 31.739 0.415 0.678 -49.048 75.373\n", + "sqft_basement[T.900.0] -9.3750 13.786 -0.680 0.496 -36.396 17.646\n", + "sqft_basement[T.906.0] 174.7209 142.085 1.230 0.219 -103.777 453.218\n", + "sqft_basement[T.910.0] 24.4157 18.227 1.340 0.180 -11.311 60.143\n", + "sqft_basement[T.915.0] 368.0925 142.165 2.589 0.010 89.439 646.746\n", + "sqft_basement[T.920.0] -25.0842 18.817 -1.333 0.183 -61.967 11.798\n", + "sqft_basement[T.930.0] 44.5215 23.238 1.916 0.055 -1.026 90.069\n", + "sqft_basement[T.935.0] 254.8604 142.122 1.793 0.073 -23.709 533.429\n", + "sqft_basement[T.940.0] -2.4900 18.217 -0.137 0.891 -38.196 33.216\n", + "sqft_basement[T.946.0] -286.9125 142.134 -2.019 0.044 -565.506 -8.319\n", + "sqft_basement[T.950.0] 27.4077 19.477 1.407 0.159 -10.769 65.584\n", + "sqft_basement[T.960.0] -2.5252 18.939 -0.133 0.894 -39.647 34.597\n", + "sqft_basement[T.970.0] 27.3292 22.521 1.214 0.225 -16.814 71.472\n", + "sqft_basement[T.980.0] -28.2232 20.048 -1.408 0.159 -67.519 11.073\n", + "sqft_basement[T.990.0] 38.2189 20.867 1.832 0.067 -2.683 79.121\n", + "bathrooms 14.2354 2.428 5.863 0.000 9.476 18.995\n", + "bedrooms -9.6075 1.562 -6.152 0.000 -12.669 -6.546\n", + "condition 9.8136 1.584 6.194 0.000 6.708 12.919\n", + "floors 20.4121 2.798 7.294 0.000 14.927 25.897\n", + "sqft_above -0.0141 0.003 -4.178 0.000 -0.021 -0.007\n", + "sqft_living 0.0825 0.003 25.611 0.000 0.076 0.089\n", + "sqft_living15 0.0542 0.003 21.066 0.000 0.049 0.059\n", + "sqft_lot -0.0011 0.001 -2.007 0.045 -0.002 -2.49e-05\n", + "sqft_lot15 -0.0050 0.001 -7.755 0.000 -0.006 -0.004\n", + "view 7.2092 1.565 4.607 0.000 4.142 10.277\n", + "yr_built -2.1008 0.047 -45.004 0.000 -2.192 -2.009\n", + "yr_renovated -0.0067 0.029 -0.233 0.816 -0.063 0.050\n", + "==============================================================================\n", + "Omnibus: 1011.610 Durbin-Watson: 1.967\n", + "Prob(Omnibus): 0.000 Jarque-Bera (JB): 1377.424\n", + "Skew: 0.460 Prob(JB): 7.88e-300\n", + "Kurtosis: 3.829 Cond. No. 2.29e+06\n", + "==============================================================================\n", + "\n", + "Notes:\n", + "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n", + "[2] The condition number is large, 2.29e+06. This might indicate that there are\n", + "strong multicollinearity or other numerical problems.\n" + ] + } + ], + "source": [ + "multi_model = model_all_independent()\n", + "print(multi_model.summary())" + ] + }, + { + "cell_type": "code", + "execution_count": 92, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "model_qq_plot(multi_model)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Observations\n", + "- The year renovated variable and many of the sqft_basement dummy variables are not statistically significant. As a result we will drop the columns.\n", + "- As a whole the model explains the variance in price much better than the simple regression model (~51% as opposed to ~33%)" + ] + }, + { + "cell_type": "code", + "execution_count": 310, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
pricebedroomsbathroomssqft_livingsqft_lotfloorsviewconditiongradesqft_aboveyr_builtsqft_living15sqft_lot15
0221.93.01.001180.05650.01.00071180.01955.01340.05650.0
1538.03.02.252570.07242.02.00072170.01951.01690.07639.0
2180.02.01.00770.010000.01.0006770.01933.02720.08062.0
3604.04.03.001960.05000.01.00271050.01965.01360.05000.0
4510.03.02.001680.08080.01.00081680.01987.01800.07503.0
\n", + "
" + ], + "text/plain": [ + " price bedrooms bathrooms sqft_living sqft_lot floors view condition \\\n", + "0 221.9 3.0 1.00 1180.0 5650.0 1.0 0 0 \n", + "1 538.0 3.0 2.25 2570.0 7242.0 2.0 0 0 \n", + "2 180.0 2.0 1.00 770.0 10000.0 1.0 0 0 \n", + "3 604.0 4.0 3.00 1960.0 5000.0 1.0 0 2 \n", + "4 510.0 3.0 2.00 1680.0 8080.0 1.0 0 0 \n", + "\n", + " grade sqft_above yr_built sqft_living15 sqft_lot15 \n", + "0 7 1180.0 1955.0 1340.0 5650.0 \n", + "1 7 2170.0 1951.0 1690.0 7639.0 \n", + "2 6 770.0 1933.0 2720.0 8062.0 \n", + "3 7 1050.0 1965.0 1360.0 5000.0 \n", + "4 8 1680.0 1987.0 1800.0 7503.0 " + ] + }, + "execution_count": 310, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "to_drop = [\"yr_renovated\", \"sqft_basement\"]\n", + "modeling_df.drop(columns=to_drop, inplace=True)\n", + "modeling_df.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 311, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " OLS Regression Results \n", + "==============================================================================\n", + "Dep. Variable: price R-squared: 0.505\n", + "Model: OLS Adj. R-squared: 0.505\n", + "Method: Least Squares F-statistic: 1047.\n", + "Date: Thu, 02 May 2024 Prob (F-statistic): 0.00\n", + "Time: 02:43:08 Log-Likelihood: -1.3735e+05\n", + "No. Observations: 21534 AIC: 2.748e+05\n", + "Df Residuals: 21512 BIC: 2.749e+05\n", + "Df Model: 21 \n", + "Covariance Type: nonrobust \n", + "=================================================================================\n", + " coef std err t P>|t| [0.025 0.975]\n", + "---------------------------------------------------------------------------------\n", + "Intercept 4610.3186 88.506 52.091 0.000 4436.841 4783.796\n", + "grade[T.11] -8.2014 8.419 -0.974 0.330 -24.704 8.301\n", + "grade[T.12] -67.4007 16.024 -4.206 0.000 -98.810 -35.992\n", + "grade[T.13] -112.8214 39.943 -2.825 0.005 -191.113 -34.530\n", + "grade[T.3] -316.1245 142.803 -2.214 0.027 -596.029 -36.220\n", + "grade[T.4] -372.7569 28.247 -13.196 0.000 -428.124 -317.390\n", + "grade[T.5] -360.0664 11.159 -32.266 0.000 -381.940 -338.193\n", + "grade[T.6] -312.2991 6.875 -45.424 0.000 -325.775 -298.823\n", + "grade[T.7] -225.0598 5.582 -40.319 0.000 -236.001 -214.119\n", + "grade[T.8] -131.0387 5.063 -25.881 0.000 -140.963 -121.115\n", + "grade[T.9] -32.3124 5.137 -6.290 0.000 -42.382 -22.243\n", + "bathrooms 16.4702 2.350 7.009 0.000 11.865 21.076\n", + "bedrooms -9.6966 1.538 -6.306 0.000 -12.711 -6.683\n", + "condition 9.1952 1.573 5.844 0.000 6.111 12.279\n", + "floors 19.7921 2.689 7.360 0.000 14.521 25.063\n", + "sqft_above -0.0162 0.003 -5.886 0.000 -0.022 -0.011\n", + "sqft_living 0.0802 0.003 30.260 0.000 0.075 0.085\n", + "sqft_living15 0.0542 0.003 21.246 0.000 0.049 0.059\n", + "sqft_lot -0.0012 0.001 -2.341 0.019 -0.002 -0.000\n", + "sqft_lot15 -0.0052 0.001 -8.035 0.000 -0.006 -0.004\n", + "view 6.8270 1.504 4.539 0.000 3.879 9.775\n", + "yr_built -2.1232 0.045 -46.756 0.000 -2.212 -2.034\n", + "==============================================================================\n", + "Omnibus: 1008.969 Durbin-Watson: 1.967\n", + "Prob(Omnibus): 0.000 Jarque-Bera (JB): 1344.377\n", + "Skew: 0.467 Prob(JB): 1.18e-292\n", + "Kurtosis: 3.791 Cond. No. 1.92e+06\n", + "==============================================================================\n", + "\n", + "Notes:\n", + "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n", + "[2] The condition number is large, 1.92e+06. This might indicate that there are\n", + "strong multicollinearity or other numerical problems.\n" + ] + } + ], + "source": [ + "multi_model = model_all_independent()\n", + "print(multi_model.summary())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Observations\n", + "- Dropping the variables did not impede the models ability to explain the variance in price.\n", + "- The models f statistic rose significantly." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "##### Price Increase Factors\n", + "The biggest contributors to a high price house are:\n", + "- Number of floors with added value of $19,792.10 for each additional floor\n", + "- Number of bathrooms which adds an additional $16,470.20 for each bathroom\n", + "- Condition of the house which contributes $9,195.20 for every increase in ranking" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Conslusions" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The following are the conlusions we derived from the analysis conducted:\n", + "\n", + "Model Effectiveness: The S.L.R.M explains only about 33% of the variation in house prices while the MLRM explaining approximately 51% of the variance in house prices.\n", + "\n", + "Significant Variables: intercept and square footage living area coefficients in the SLRM are statistically significant. In the MLRM variables such as the number of floors, number of bathrooms, and condition of the house are statistically significant contributors to house prices.\n", + "\n", + "Price Increase Factors: Each floor adds approximately \n", + "$16,470.20. Improving the condition of the house can also lead to higher prices, with each increase in ranking contributing approximately $9,195.20." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Recommendations" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Improve the predictive power of the regression models by including variables that are statistically significant and dropping those that are not.\n", + "\n", + "Pay close attention to factors such as the number of floors, number of bathrooms, and condition of the house, as these variables have a significant impact on house prices.\n", + "\n", + "Consider collecting additional data or exploring alternative modeling techniques to further enhance the accuracy of house price predictions." ] } ], @@ -40,7 +4021,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.6.4" + "version": "3.11.5" } }, "nbformat": 4,