Logging evaluation metrics

Tracking performance metrics allows you to monitor degradations, and you can make decisions about when to update your model to maintain a high level of accuracy. You decide you will log metrics after your model finishes an evaluation loop.

Some data has been pre-loaded:

accelerator is an instance of Accelerator
eval_metric is a dictionary of metrics like accuracy and f1
num_epochs is the number of epochs

Deze oefening maakt deel uit van de cursus

Efficient AI Model Training with PyTorch

Cursus bekijken

Oefeninstructies

Call a method to log evaluation metrics of the model.
Log "accuracy" and "f1" score as evaluation metrics.
Track the epoch number using epoch of the training loop.

Praktische interactieve oefening

Probeer deze oefening eens door deze voorbeeldcode in te vullen.

accelerator = Accelerator(project_dir=".", log_with="all")
accelerator.init_trackers("my_project")

for epoch in range(num_epochs):
    # Training loop is here
    # Evaluation loop is here
    # Call a method to log metrics
    ____.____({
        # Log accuracy and F1 score as metrics
        "accuracy": ____["accuracy"],
        "f1": ____["f1"],
    # Track the epoch number
    }, ____=____)

accelerator.end_training()

Code bewerken en uitvoeren