LuciferML a Semi-Automated Machine Learning Library by d4rk-lucif3r

ReadTheDocs

About

The LuciferML is a Semi-Automated Machine Learning Python Library that works with tabular data. It is designed to save time while doing data analysis. It will help you right from data preprocessing to Data Prediction.

The LuciferML will help you with

Preprocessing Data:
- Encoding
- Splitting
- Scaling
- Dimensionality Reduction
- Resampling
Trying many different machine learning models with hyperparameter tuning,

Installation

pip install lucifer-ml

Available Preprocessing Techniques

Skewness Correction

Takes Pandas Dataframe as input. Transforms each column in dataset except the columns given as an optional parameter. Returns Transformed Data.

Example:

All Columns:

 from luciferml.preprocessing import Preprocess as prep
 import pandas as pd
 dataset = pd.read_csv('/examples/Social_Network_Ads.csv')
 dataset = prep.skewcorrect(dataset)

Except column/columns:

 from luciferml.preprocessing import Preprocess as prep
 import pandas as pd
 dataset = pd.read_csv('/examples/Social_Network_Ads.csv')
 dataset = prep.skewcorrect(dataset,except_columns=['Purchased'])

More about Preprocessing here

Available Modelling Techniques

Classification

Available Models for Classification

 - 'lr'  : 'Logistic Regression',
 - 'sgd' : 'Stochastic Gradient Descent',
 - 'perc': 'Perceptron',
 - 'pass': 'Passive Aggressive Classifier',
 - 'ridg': 'Ridge Classifier', 
 - 'svm' : 'Support Vector Machine',
 - 'knn' : 'K-Nearest Neighbours',
 - 'dt'  : 'Decision Trees',
 - 'nb'  : 'Naive Bayes',
 - 'rfc' : 'Random Forest Classifier',
 - 'gbc' : 'Gradient Boosting Classifier',
 - 'ada' : 'AdaBoost Classifier',
 - 'bag' : 'Bagging Classifier',
 - 'extc': 'Extra Trees Classifier',
 - 'lgbm': 'LightGBM Classifier',
 - 'cat' : 'CatBoost Classifier',
 - 'xgb' : 'XGBoost Classifier',
 - 'ann' : 'Multilayer Perceptron Classifier',
 - 'all' : 'Applies all above classifiers'

Example:

 from luciferml.supervised.classification import Classification
 dataset = pd.read_csv('Social_Network_Ads.csv')
 X = dataset.iloc[:, :-1]
 y = dataset.iloc[:, -1]
 classifier = Classification(predictor = ['lr'])
 classifier.fit(X, y)
 result = classifier.result()

More About Classification

Regression

Available Models for Regression

 - 'lin' : 'Linear Regression',
 - 'sgd' : 'Stochastic Gradient Descent Regressor',
 - 'elas': 'Elastic Net Regressot',
 - 'krr' : 'Kernel Ridge Regressor',
 - 'br'  : 'Bayesian Ridge Regressor',
 - 'svr' : 'Support Vector Regressor',
 - 'knr' : 'K-Nearest Regressor',
 - 'dt'  : 'Decision Trees',
 - 'rfr' : 'Random Forest Regressor',
 - 'gbr' : 'Gradient Boost Regressor',
 - 'ada' : 'AdaBoost Regressor',
 - 'bag' : 'Bagging Regressor',
 - 'extr': 'Extra Trees Regressor',
 - 'lgbm': 'LightGBM Regressor',
 - 'xgb' : 'XGBoost Regressor',
 - 'cat' : 'Catboost Regressor',
 - 'ann' : 'Multilayer Perceptron Regressor',
 - 'all' : 'Applies all above regressors'

Example:

 from luciferml.supervised.regression import Regression
 dataset = pd.read_excel('examples\Folds5x2_pp.xlsx')
 X = dataset.iloc[:, :-1]
 y = dataset.iloc[:, -1]
 regressor = Regression(predictor = ['lin'])
 regressor.fit(X, y)
 result = regressor.result()

More about Regression here

Hyperparameter Tuning

LuciferML is powered by Optuna for Hyperparam tuning. Just add "tune = True" in either Regressor or Classifier it will start tuning the model/s with Optuna.

Persistence

LuciferML's model can be saved as a pickle file. It will save both the model and the scaler to the pickle file.

- Saving

    Ex: 
        regressor.save([<path-to-model.pkl>, <path-to-scaler.pkl>])

A new LuciferML Object can be loaded as well by specifying path of model and scaler

- Loading

    Ex: 
        regressor = Regression(path = [<path-to-model.pkl>, <path-to-scaler.pkl>])

These are applicable for both Classification and Regression.

Examples

Please refer to more examples here

To-Do's

Metadata

Release files for lucifer-ml 0.0.79

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for lucifer-ml 0.0.79
File	Size	Uploaded
lucifer-ml-0.0.79.tar.gz	34.1 kB	Jan 14, 2022	Details

Built distribution (wheel)

Table of built distributions (wheels) for lucifer-ml 0.0.79
File	Interpreter	ABI	Platform	Reset
lucifer_ml-0.0.79-py3-none-any.whl 43.5 kB Jan 14, 2022	Python 3	none	any	Details

Total release size: 77.6 kB

Release files / lucifer-ml-0.0.79.tar.gz

Download URL	lucifer-ml-0.0.79.tar.gz
Size	34.1 kB
Tags	Source
SHA-256 checksum How to use checksums	`23a95396938b442c6cf399582b15cfdbf3caf33b8a0fa395441e69040bc819bf`
BLAKE2b-256 checksum How to use checksums	`22017b5203793d9cc5e1c962ab0a8fcab0394ca35e59d7558cd97f35815952cc`
Upload date	Jan 14, 2022
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/3.7.1 importlib_metadata/4.10.0 pkginfo/1.8.2 requests/2.27.1 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.10.1`

Release files / lucifer_ml-0.0.79-py3-none-any.whl

Download URL	lucifer_ml-0.0.79-py3-none-any.whl
Size	43.5 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`5ea9b22d34c8f373df47c9fcd0e0121318d434d98ade48f7b9d954332d63d127`
BLAKE2b-256 checksum How to use checksums	`4a128b6d9dffc67877e031c6df8d6f8d3c5e52a40457d4301800c151a98d83b4`
Upload date	Jan 14, 2022
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/3.7.1 importlib_metadata/4.10.0 pkginfo/1.8.2 requests/2.27.1 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.10.1`