GML 3.0.0

Automating Data Science

GML!

Tired of doing Data Science manually? GML is here for you!

GML is an automatic data science library in python built on top of multiple Python packages. Complete features which we offer are listed as:

Installation:

pip install GML

https://pypi.org/project/GML

Features:

Auto Feature Engineering

from GML import FeatureEngineering

fe = FeatureEngineering(Data, 'target', fill_missing_data=True, encode_data=True, 
                        normalize=True, remove_outliers=True, 
                        new_features=True, feateng_steps=2 ) # feateng_steps = 0 for features selection without feature creation

X_new, y, test = fe.get_new_data()

Click Here for complete DEMO

Auto EDA (Powered by Sweetviz)

from GML import sweetviz

result1 = sweetviz.compare([train,'train'],[test,'test'],'target') 
result2 = sweetviz.analyze([train,'train'])

result.show_html()
result2.show_html()

Click Here for complete DEMO

Auto Machine Learning

from GML import AutoML

gml_ml = AutoML()

gml_ml.GMLClassifier(X, y, metric = accuracy_score, folds = 10)

Click Here for complete DEMO

Auto Text Cleaning

from GML import AutoNLP

nlp = AutoNLP()

cleanX = X.apply(lambda x: nlp.clean(x))

Click Here for complete DEMO

Auto Text Classification using transformers

from GML import AutoNLP

nlp = AutoNLP()

nlp.set_params(cleanX, tokenizer_name='roberta-large-mnli', BATCH_SIZE=4,
               model_name='roberta-large-mnli', MAX_LEN=200)

model = nlp.train_model(tokenizedX, y)

Click Here for complete DEMO

Auto Text Classification using transformers

from GML import AutoNLP

nlp = AutoNLP()

nlp.set_params(cleanX, tokenizer_name='roberta-large-mnli', BATCH_SIZE=4,
               model_name='roberta-large-mnli', MAX_LEN=200)

model = nlp.train_model(tokenizedX, y)

Click Here for complete DEMO

Auto Image Classification with Augmentation

from GML import Auto_Image_Processing

gml_image_processing = Auto_Image_Processing()

model = gml_image_processing.imgClassificationcsv(img_path = './covid_image_data/train', 
                                                  train_path = './covid_image_data/Training_set_covid.csv', 
                                                  model_list = models,
                                                 tfms = True, advance_augmentation = True, 
                                                  epochs=1)

Click Here for complete DEMO

Text Augmentation using transformers: GPT-2

from GML import AutoNLP

nlp = AutoNLP()

nlp.augmentation_train('./data.csv')

nlp.set_params(X['Text'])

new_Text = nlp.augmentation_generate(y = y, SENTENCES = 100) 

Click Here for complete DEMO

More cool features and handling of different data types like audio data etc will be added in future.
Feel free to give suggestions, report bugs and contribute.