sagnik1511 / Tabular-AutoML

Python Auto-ML Package for Tabular Datasets
MIT License
23 stars 13 forks source link
automl modular-architecture numpy pandas python-package python3 scikit-learn
Tabular-AutoML

Tabular-AutoML

AutoML Package for tabular datasets

Tabular dataset tuning is now hassle free!

Run one liner command and get best tuning and processed dataset in a go.

Python Git

Used Python Libraries :
lightgbm numpy numpy numpy

Installation & Usage


  1. Create a Virtual Environment : Tutorial
  2. Clone the repository.
  3. Open the directory with cmd.
  4. Copy this command in terminal to install dependencies.
    pip install -r requirements.txt
  5. Installing the requirements.txt may generate some error due to outdated MS Visual C++ Build. You can fix this problem using this.
  6. First check the parser variable that has to be passed with all customizations.
    
    >>> python -m tab_automl.main --help
    usage: main.py [-h] -d  -t  -tf  [-p] [-f] [-spd] [-sfd] [-sm]

automl hyper parameters

optional arguments: -h, --help show this help message and exit -d , --data-source File path -t , --problem-type Problem Type , currently supporting regression or classification -tf , --target-feature Target feature inside the data -p , --pre-proc If data processing is required -f , --fet-eng If feature engineering is required -spd , --save-proc-data Save the processed data -sfd , --save-fet-data Save the feature engineered data -sm , --save-model Save the best trained model

7. Now run the command with your custom data, problem type and target feature
```python
>>> # For Regression Problem
>>> python -m tab_automl.main -d "your custom data scource\custom_data.csv" -t "regression" -tf "your_custom_target_feature" -spd "true" -sfd "true" -sm "true"

>>> # For Classification Problem
>>> python -m tab_automl.main -d "your custom data scource\custom_data.csv" -t "classification" -tf "your_custom_target_feature" -spd "true" -sfd "true" -sm "true"

Contributing Guidelines


  1. Comment on the issue on which you want to work.
  2. If you get assigned, fork the repository.
  3. Create a new branch which should be named on your GitHub user_id , e.g. sagnik1511.
  4. Update the changes on that branch.
  5. Create a PR (Pull request) to the JWOC branch of the parent repository.
  6. The PR title should be named like this [Issue {Issue Number}] Heading of the issue.
  7. Describe the changes you have done with proper reasons.

This branch will hold all updates of JWOC.

Contributors


  1. Sagnik Roy : sagnik1511

If you like the project, do ⭐

Also follow me on GitHub , Kaggle , LinkedIn

Thank You for Visiting :)