mhpp 0.3.0

A median house price prediction project to demonstrate packaging.

Introduction

A median house price prediction (mhpp) project is dealt to demonstrate packaging. This package helps to fetch, preprocess and split the data set into train and test and also train models using Linear Regression, Decision Tree and Random Forest algorithms and also evaluate them using MSE (Mean Squarred Error), RMSE (Root mean square error) and MAS (Mean Absolute Error).

Installation

Create a virtual environment using conda before installation.

  conda env create -f deploy/conda/env.yaml

Using pip

  pip install mhpp

From source

1. Clone the repo using git clone command. Note: Requires colab access as of now.

    git clone [e-mail]:varun-mle/mle-training.git

2. Execute the below command.

    pip install . 

Usage

1. For fetching data fetch-data command is used. Below command helps to know its usage

       fetch-data -h
   

2. For preprocessing and splitting the data into train & test datasets train-test-data is used. Below command helps to know its usage.

       train-test-data -h 
   

3. For training the model train command is used. Below command helps to know its usage

       train -h 
   

4. For evaluating the model evaluate command is used. Below command helps to know its usage

       evaluate -h
   

Refer docs folder for the detailed documentation.