Type multi_worker_training_state
Namespace tensorflow.python.keras.distribute.multi_worker_training_state
Methods
Properties
- checkpoint_exists_fn
- CKPT_SAVED_EPOCH_UNUSED_VALUE
- CKPT_SAVED_EPOCH_UNUSED_VALUE_dyn
- MultiWorkerTrainingState_fn
- remove_checkpoint_if_exists_fn
Fields
Public static methods
bool checkpoint_exists(object filepath)
object checkpoint_exists_dyn(object filepath)
Checks whether a V1 or V2 checkpoint exists with the specified prefix. (deprecated) Warning: THIS FUNCTION IS DEPRECATED. It will be removed in a future version.
Instructions for updating:
Use standard file APIs to check for files with this prefix. This is the recommended way to check if a checkpoint exists, since it takes
into account the naming difference between V1 and V2 formats.
Returns
-
object - A bool, true if a checkpoint referred to by `checkpoint_prefix` exists.
bool remove_checkpoint_if_exists(object ckpt_dir, object filepath)
object remove_checkpoint_if_exists_dyn(object ckpt_dir, object filepath)
Public properties
PythonFunctionContainer checkpoint_exists_fn get;
int CKPT_SAVED_EPOCH_UNUSED_VALUE get; set;
object CKPT_SAVED_EPOCH_UNUSED_VALUE_dyn get; set;
PythonFunctionContainer MultiWorkerTrainingState_fn get;
PythonFunctionContainer remove_checkpoint_if_exists_fn get;
Public fields
string CKPT_SAVED_EPOCH
return string
|