crested.tl.TaskConfig

class crested.tl.TaskConfig(optimizer: keras.optimizers.Optimizer, loss: keras.losses.Loss, metrics: list[keras.metrics.Metric])

Task configuration (optimizer, loss, and metrics) for use in tl.Crested.

The TaskConfig class is a simple NamedTuple that holds the optimizer, loss, and metrics

Parameters:

• optimizer – Optimizer used for training.

• loss – Loss function used for training.

• metrics – Metrics used for training.

Example

>>> optimizer = tf.keras.optimizers.Adam(learning_rate=1e-3)
>>> loss = tf.keras.losses.BinaryCrossentropy(from_logits=False)
>>> metrics = [
...     tf.keras.metrics.AUC(
...         num_thresholds=200,
...         curve="ROC",
...     )
... ]
>>> configs = TaskConfig(optimizer, loss, metrics)

See also

crested.tl.default_configs

Attributes table

loss	Alias for field number 1
metrics	Alias for field number 2
optimizer	Alias for field number 0

Methods table

count(value, /)	Return number of occurrences of value.
index(value[, start, stop])	Return first index of value.
to_dict()	Convert the TaskConfig to a dictionary.

Attributes

TaskConfig.loss: Loss

Alias for field number 1

TaskConfig.metrics: list[Metric]

Alias for field number 2

TaskConfig.optimizer: Optimizer

Alias for field number 0

Methods

TaskConfig.count(value, /)

Return number of occurrences of value.

TaskConfig.index(value, start=0, stop=9223372036854775807, /)

Return first index of value.

Raises ValueError if the value is not present.

TaskConfig.to_dict()

Convert the TaskConfig to a dictionary.

Useful for logging and saving the configuration.

Return type:

dict

Returns:

Dictionary representation of the TaskConfig.