AutoKeras

The autokeras_imdb_example.py example script demonstrates the integration of Trains into code which uses autokeras. It trains text classification networks on the Keras built-in IMDB dataset, using the autokeras TextClassifier class, and searches for the best model. It uses two TensorBoard callbacks, one for training and one for testing. Trains automatically logs everything the code sends to TensorBoard. When the script runs, it creates an experiment named autokeras imdb example with scalars, which is associated with the autokeras project.

Scalars

Loss and the accuracy metric per iteration which appear in the RESULTS tab, SCALARS tab, along with the resource utilization plots as :monitor: machine.

Hyperparameters

The TensorFlow DEFINES, which are automatically logged when tensorflow is used, appear in the HYPER PARAMETERS tab.

Log

Text printed to the console for training progress, as well as all other console output, appear in the RESULTS tab, LOG tab.

Artifacts

Trains tracks the input and output model with the experiment, but the Trains Web (UI) shows the model details separately.

Input model

In the experiment details, ARTIFACTS tab, Input Model area, you can see Trains logging the:

  • Model name
  • Experiment creating the model
  • Model configuration

In the model details (which appear when you click the model name, expand image above), you can see the following:

  • Model location (URL), checkpoint models (URLs), experiment creating the model, and other general information about the model.
  • The model configuration.

Output model

In the Output Model area, you can see the:

  • Model name
  • Output model configuration.

In the model details, you can see the following:

  • Model location (URL), checkpoint models (URLs), experiment creating the model, and other general information about the model.
  • The model configuration.