Loan Prediction Model
Project description
Packaging the ML Model of Classification
Problem Statement
- Company wants to automate the loan eligibility process based on customer detail provided while filling online application form.
- It is a classification problem where we have to predict whether a loan would be approved or not.
Data
The data corresponds to a set of financial requests associated with individuals.
Variables | Description |
---|---|
Loan_ID | Unique Loan ID |
Gender | Male/ Female |
Married | Applicant married (Y/N) |
Dependents | Number of dependents |
Education | Applicant Education (Graduate/ Under Graduate) |
Self_Employed | Self employed (Y/N) |
ApplicantIncome | Applicant income |
CoapplicantIncome | Coapplicant income |
LoanAmount | Loan amount in thousands |
Loan_Amount_Term | Term of loan in months |
Credit_History | credit history meets guidelines |
Property_Area | Urban/ Semi Urban/ Rural |
Loan_Status | Loan approved (Y/N) |
Source: Kaggle
Running Locally
Add PYTHONPATH variable for ~/.bash_profile
for MacOS
</code></pre>
<h2>Virtual Environment</h2>
<p>Install virtualenv</p>
<pre lang="python"><code>python3 -m pip install virtualenv
Check version
virtualenv --version
Create virtual environment
virtualenv ml_package
Activate virtual environment
For Linux/Mac
source ml_package/bin/activate
For Windows
ml_package\Scripts\activate
Deactivate virtual environment
deactivate
Directory structure
prediction_model
├── MANIFEST.in
├── prediction_model
│ ├── config
│ │ ├── config.py
│ │ └── __init__.py
│ ├── datasets
│ │ ├── __init__.py
│ │ ├── test.csv
│ │ └── train.csv
│ ├── __init__.py
│ ├── pipeline.py
│ ├── predict.py
│ ├── processing
│ │ ├── data_handling.py
│ │ ├── __init__.py
│ │ └── preprocessing.py
│ ├── trained_models
│ │ ├── classification.pkl
│ │ └── __init__.py
│ ├── training_pipeline.py
│ └── VERSION
├── README.md
├── requirements.txt
├── setup.py
└── tests
├── pytest.ini
└── test_prediction.py
Build the Package
-
Goto Project directory and install dependencies
pip install -r requirements.txt
-
Create Pickle file after training:
python prediction_model/training_pipeline.py
-
Create source distribution and wheel
python setup.py sdist bdist_wheel
Installation of Package
Go to project directory where setup.py
file is located
- To install it in editable or developer mode
pip install -e .
.
refers to current directory
-e
refers to --editable mode
- Normal installation
pip install .
.
refers to current directory
- Also can be installed from git as well after pushing to github
pip install git+https://github.com/manifoldailearning/prediction_model.git
Testing the Package Working
- Remove the PYTHONPATH from environment variables
- Goto a separate location which is outside of package directory
- Create a new virual environment using the commands mentioned above & activate it
- Before installing, test whether you are able to import the package of
prediction_model
- (you should not be able to do it) - Now in the new environment install the package from github
pip install git+https://github.com/manifoldailearning/prediction_model.git
- Now try importing the prediction_model, you should be able to do it successfully
- Extras : Run training pipeline using the package, and also conduct the test
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Hashes for prediction_model_manifoldailearning-1.0.0.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | fbbad14abe3dff7d120d65c96725be9b72545f2b2e966f93a6bed7fd7e893cee |
|
MD5 | 0c7fba4d97704546a9333ecd5a06a8be |
|
BLAKE2b-256 | 050cec5cea4cd83dd28791951985824a88e9e65b03857658a957f61f885ad217 |