Configuring Models

Trains is now ClearML

This documentation applies to the legacy Trains versions. For the latest documentation, see ClearML.

The model_config.py example demonstrates configuring a model and defining class label enumeration. Connect the configuration and class label enumeration to a Task and, once connected, Trains tracks any changes to them. When Trains stores a model, in any framework, Trains stores the configuration and class label enumeration with it.

When the script runs, it creates an experiment named Model configuration example, which is associated with the examples project.

Configuring models

Using a configuration file

Connect a configuration file to a Task by reading the file, and then calling the Task.connect_configuration method with the file.

# Connect a local configuration file
config_file = os.path.join('data_samples', 'sample.json')
config_file = task.connect_configuration(config_file)

Trains reports the configuration in the Trains Web (UI), experiment details, ARTIFACTS tab, Input Model area. See the image in the next section.

Configuration dictionary

Connect a configuration dictionary to a Task by creating a dictionary, and then calling the Task.connect_configuration method with the dictionary. After the configuration is connected, Trains tracks changes to it.

model_config_dict = {
    'value': 13.37,
    'dict': {'sub_value': 'string', 'sub_integer': 11},
    'list_of_ints': [1, 2, 3, 4],
}
model_config_dict = task.connect_configuration(model_config_dict)

# We now update the dictionary after connecting it, and the changes will be tracked as well.
model_config_dict['new value'] = 10
model_config_dict['value'] *= model_config_dict['new value']

Trains reports the configuration in the Trains Web (UI), experiment details, ARTIFACTS tab, Input Model area.

Class label enumeration

Connect a class enumeration dictionary by creating the dictionary, and then calling the Task.connect_label_enumeration method with the dictionary.

# store the label enumeration of the training model
labels = {'background': 0, 'cat': 1, 'dog': 2}
task.connect_label_enumeration(labels)

Trains reports the configuration in the Trains Web (UI), models details, LABELS tab.

This example does not store a model, however.