Skip to content

Latest commit

 

History

History
37 lines (26 loc) · 1.18 KB

Monitoring.md

File metadata and controls

37 lines (26 loc) · 1.18 KB
Raw

Monitoring

There are several options for monitoring the experiment.

# turn off progress bar
Scan(disable_progress_bar=True)

# enable live training plot
from talos.callbacks import TrainingPlot

out = model.fit(X,
                Y,
                epochs=20,
                callbacks=[TrainingPlot()])

# turn on parameter printing
Scan(print_params=True)

Progress Bar : A round-by-round updating progress bar that shows the remaining rounds, together with a time estimate to completion. Progress bar is on by default.

Live Monitoring : Live monitoring provides an epoch-by-epoch updating line graph that is enabled through the TrainingPlot() custom callback.

Round Hyperparameters : Displays the hyperparameters for each permutation. Does not work together with live monitoring.

Local Monitoring

Epoch-by-epoch training data is available during the experiment using the ExperimentLog:

model.fit(...
          callbacks=[talos.callbacks.ExperimentLog('experiment_name', params)])

Here params is the params dictionary in the Scan() input model. Both experiment_name and experiment_id should match with the current experiment, as otherwise