Skip to content

Visualization of PyCaffe Experiment Hyperparameters

This topic shows you how to set experiment hyperparams and their effects. The topic builds on the script that resulted from steps in Getting Started for PyCaffe with Solve.

The steps that are covered are:

  • Define a hyperparam.
  • Set hyperparams to MissingLink's callback.

Note

For each framework that MissingLink supports, there are hyperparameters that will be retrieved automatically. See the list.

Preparation

Go through Getting Started for PyCaffe with Solve.

Note

Ensure that you can successfully run the mnist.py training script that resulted from integration with the MissingLink SDK. In the steps that follow below, the script is further developed to include hyperparams.

Write code

  1. Add a dropout rate hyperparam:

    DROPOUT_RATE = 0.1
    
  2. Set hyperparams to an experiment:

    In the base script, just after the display name and description definition of an experiment, add an extra line right below to set some hyperparams.

    missinglink_callback.set_hyperparams(
        dropout_rate=DROPOUT_RATE)
    

You should have added hyperparams to your experiment successfully.

  • Inspect the resulting script here.
  • Run the new script and see how MissingLink's dashboard helps with monitoring the experiment's hyperparams. A description follows.

Viewing the new functionality on the dashboard

You can see the hyperparams across different experiments on your MissingLink dashboard.

Hyperparameters that are retrieved automatically

The following hyperparameters are retrieved automatically if they are defined, regardless of whether the set_hyperparams function is used to set them:

  • Algorithm
  • Gamma
  • Learning rate
  • Learning rate policy
  • Momentum
  • Weight decay
  • Batch size
  • Total batches
  • Epoch size
  • Total epochs
  • Max iterations