Metadata-Version: 2.1
Name: training_dashboard
Version: 0.1.0
Summary: No bullshit, dead simple training visualizer for tf-keras.
Home-page: https://github.com/vibhuagrawal14/training_dashboard
Author: Vibhu Agrawal
Author-email: vibhu.agrawal14@gmail.com
License: MIT
Description: # Training Dashboard
        A no bullshit, dead simple training visualizer for tf-keras.
        
        
        ## Usage
        
        TrainingDashboard is meant to be used as a callback passed to the fit() function.
        
        ```python
          from training_dashboard import TrainingDashboard
          callback = TrainingDashboard(validation=True,
                                       min_loss=0,
                                       metrics=["accuracy", "auc"],
                                       batch_step=10,
                                       min_metric_dict={"accuracy": 0, "auc": 0},
                                       max_metric_dict={"accuracy": 1, "auc": 1})
          model.fit(x_train,
                    y_train,
                    batch_size=512,
                    epochs=25,
                    verbose=1,
                    validation_split=0.2,
                    callbacks=[callback])
        ```
        
        ## Example Output
        
        <img src="https://i.imgur.com/D91YmwT.gif" width="700"/>
        
        
Platform: UNKNOWN
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: Implementation :: CPython
Classifier: Programming Language :: Python :: Implementation :: PyPy
Description-Content-Type: text/markdown
