base_context¶

Primitive Context Interface.

Contexts are checkpointable (subclassed from tf.train.Checkpoint) collections of variable encapsulated in a Python Class as a way to seamlessly handle information transfer.

Classes

BaseContext

ashpy.contexts.BaseContext provide an interface for all contexts.

class ashpy.contexts.base_context.BaseContext(metrics=None, dataset=None, log_eval_mode=<LogEvalMode.TEST: 0>, global_step=<tf.Variable 'global_step:0' shape=() dtype=int64, numpy=0>, ckpt=None)[source]¶

Bases: object

ashpy.contexts.BaseContext provide an interface for all contexts.

__init__(metrics=None, dataset=None, log_eval_mode=<LogEvalMode.TEST: 0>, global_step=<tf.Variable 'global_step:0' shape=() dtype=int64, numpy=0>, ckpt=None)[source]¶

Initialize the Context.

Parameters

metrics (list of [ashpy.metrics.metric.Metric]) – List of ashpy.metrics.metric.Metric objects.
dataset (tf.data.Dataset) – The dataset to use, that contains everything needed to use the model in this context.
log_eval_mode (ashpy.modes.LogEvalMode) – Models’ mode to use when evaluating and logging.
global_step (tf.Variable) – Keeps track of the training steps.
ckpt (tf.train.Checkpoint) – Checkpoint to use to keep track of models status.

Return type

None

_validate_metrics()[source]¶: Check if every metric is an ashpy.metrics.Metric.

property dataset¶

Retrieve the dataset.

Return type: DatasetV2
Returns: tf.data.Dataset.

property global_step¶

Retrieve the global_step.

Return type: Variable
Returns: tf.Variable.

property log_eval_mode¶

Retrieve model(s) mode.

Return type: LogEvalMode
Returns: ashpy.modes.LogEvalMode.

measure_metrics()[source]¶

Measure the metrics.

Return type: None

property metrics¶

Retrieve the metrics.

Return type: List[Metric]
Returns: list of [ashpy.metrics.metric.Metric].

model_selection()[source]¶

Use the metrics to perform model selection.

Return type: None