ml_scripts

Ready to use Machine Learning Python Scripts.

Setup

First you need to install the required python packages.

pip install -r requirements.txt

Scripts

Data Loading

$ python load_data.py -h
usage: load_data.py [-h] -u URL -p PATH

optional arguments:
-h, --help            show this help message and exit
-u URL, --url URL     url of data file.
-p PATH, --path PATH  path to save the file.

Data Processessing

$ python process_data.py -h
usage: process_data.py [-h] -i INPUT_DATA -o OUTPUT_DATA

optional arguments:
-h, --help            show this help message and exit
-i INPUT_DATA, --input_data INPUT_DATA
                        input data file.
-o OUTPUT_DATA, --output_data OUTPUT_DATA
                        output preprocessed data.

Data Preparation

$ python prepare_data.py -h 
usage: prepare_data.py [-h] -i INPUT_DATA -p PATH

optional arguments:
-h, --help            show this help message and exit
-i INPUT_DATA, --input_data INPUT_DATA
                        input data file.
-p PATH, --path PATH  path for saving saving train and test data.

Model Training

$ python train.py -h        
usage: train.py [-h] -i INPUT_DATA -m MODEL_PATH

optional arguments:
-h, --help            show this help message and exit
-i INPUT_DATA, --input_data INPUT_DATA
                        input data file.
-m MODEL_PATH, --model_path MODEL_PATH
                        path for saving trained model.

Model Scoring

$ python predict.py -h
usage: predict.py [-h] -f FEATURES -m MODEL_PATH

optional arguments:
-h, --help            show this help message and exit
-f FEATURES, --features FEATURES
                        list of feeatures delimited by ',' .
-m MODEL_PATH, --model_path MODEL_PATH
                        path for loading the model.

Model Evaluation

$ python evaluate.py -h
usage: evaluate.py [-h] -i INPUT_DATA -m MODEL_PATH

optional arguments:
-h, --help            show this help message and exit
-i INPUT_DATA, --input_data INPUT_DATA
                        input data file.
-m MODEL_PATH, --model_path MODEL_PATH
                        path for loading the model.

Use case

Load data file

$ python load_data.py -u http://archive.ics.uci.edu/ml/machine-learning-databases/wine-quality/winequality-red.csv -p data/raw 
#INFO: Data is succefully loaded!

Process data

$ python process_data.py -i data/raw/winequality-red.csv -o data/processed/winequality-red.csv
#INFO: Data is succefully processed!

Prepare data

$ python prepare_data.py -i data/processed/winequality-red.csv -p data/prepared/
#INFO: Data is succefully prepared!

Train the ML model

$ python train.py -i data/prepared/train_winequality-red.csv -m models/elastic_net.pkl
#INFO: Model is succefully trained!

Score the model on a new instance of data

$ python predict.py -f "7.4,0.66,0.0,1.8,0.075,13.0,40.0,0.9978,3.51,0.56,9.4" -m models/elastic_net.pkl
the predicted value  is : 5.613313110911338

Evaluate the model

$ python evaluate.py -i data/prepared/test_winequality-red.csv -m models/elastic_net.pkl 
#INFO:  RMSE: 0.7560270879287759
#INFO:  MAE: 0.5986176396638518
#INFO:  R2: 0.12714137617914456
#INFO: Model is succefully evaluated!