modellogger 0.2.3

Modelloger is a Python library for storing model's profile and rapid inter model comparision.

Modelloger is a Python library for storing model’s profile and rapid inter model comparision. Powered by dash and SQLITE3, It’s compact ,light and interactive yet powerfull tool to gain usefull insights.

Documentaion :

Read the detailed documentation here

Installation

Using pip

Use the package manager pip to install modelloger.

pip install modellogger

Usage

from modellogger import ModelLogger

#initialise a modelloger instance
path = "c/path/to/db/databasename.db"
mlog = ModelLogger(path)

#If you are already using a db created by modelloger you can directly load it by stating it's path
#if you are creating a new project just give location where you want to store the db followed by a name.db

• Try It Now?

Functionalities

Now you are ready to rock and roll. Out of the box you will have the following functionalities:

store_model()

gboost = GradientBoostingClassifier()
gboost.fit(X_train,y_train)
mlog.store_model('my_model_name',gboost,X_train,1.0)
#alternatively
mlog.store_model('my_model_name',gboost,X_train,get_score(gboost),save_pickle = True)

view_results()

mlog.view_results(True,'my_report')

best_model()

mlog.best_model(smaller_the_better=True)

delete_model()

mlog.delete_model(Model_name = "Mymodel")
mlog.delete_model(Model_id = 1)

model_profiles()

mlog.model_profiles('All')
mlog.model_profiles(5)

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.

License

apache-2.0