Creator : Sathya Krishnan Suresh
This is a python package to quicken the modelling and data analysis process.
PyPi : https://pypi.org/project/mb-scripts/
-
feature selection
Feature selection sub-package has been added which consists of UnivariateFeatureSelection and CombinationFeatureSelection. -
generate_time_series
This function can be used to test out new Recurrent Neural Architectures. -
Plotting feature importances using RandomForest
This function helps selecting the features that are most correlated with the target variable with the help of RandomForest -
Decomposition
The new model in 🔥Decomposition🔥 contains standard decomposition methods that help with dimensionality reduction. -
metrics update
A lot of regression metrics have been added along with multiclass precision metrics variations.
The package can be downloaded using
pip install mb-scripts==<latest_version>
Latest version : 0.1.0.
Once you have installed mb_scripts you can begin using it.
Here are some examples for using mb_scripts.
train_validation_curve_for_rf
- used to monitor RandomForestClassifier
's overfitting
plot_decision_boundary
is used to visually look at the decision boundary of classification functions
classifiers_metrics
returns a dataframe that consists of precision, recall, accuracy_score and f1_score for all the classification models passed.
I am writing scripts regularly so the versions will keep changing for the next one month. Stay tuned.