ml_express

A Python library for day to day data analysis and machine learning. This aims to make data building, cleaning and machine learning much much faster.

Installation

pip install ml-express

Usage

import ml_express as mlx
from ml_express import eda

# this will create an html report and report will be saved in report folder in working directory
eda.create_summary_report(df)

# generate summary statistics of data type and categorical vars
eda.gen_eda(df)

# using correlation modules
# correlation for categorical vars
dfcramers = correlation.get_corr_df_cat(cat_cols,train_df_new)

# Draw the heatmap using seaborn
f, ax = plt.subplots(figsize=(12, 6))

sns.heatmap(dfcramers, annot=True, fmt='.2f',cmap='coolwarm',vmin=0 )
plt.title("Important Catg variables correlation map", fontsize=15)
plt.show()


# correlation for categorical and continuous vars
dfcramers = correlation.get_corr_df_cat_cont(cat_cols ,num_cols ,train_df_new)

# Draw the heatmap using seaborn
f, ax = plt.subplots(figsize=(12, 6))

sns.heatmap(dfcramers, annot=True, fmt='.2f',cmap='coolwarm',vmin=0,vmax = 1 )
plt.title("Important Catg vs Cont variables correlation map", fontsize=15)
plt.show()

Preprocessing

Contributing

Contributions are very welcome. Please feel free to submit PR.

License

Distributed under the terms of the MIT license, "ml-express" is free and open source software

Issues

If you encounter any problems, please file an issue along with a detailed description.