LostTech.TensorFlow : API Documentation

Type RunConfig

Namespace tensorflow.contrib.learn

Parent ClusterConfig

Interfaces IRunConfig

This class specifies the configurations for an `Estimator` run.

This class is a deprecated implementation of tf.estimator.RunConfig interface.

Methods

Properties

Public instance methods

IEnumerator<string> uid(IEnumerable<string> whitelist)

Generates a 'Unique Identifier' based on all internal fields. (experimental)

Warning: THIS FUNCTION IS EXPERIMENTAL. It may change or be removed at any time, and without warning.

Caller should use the uid string to check `RunConfig` instance integrity in one session use, but should not rely on the implementation details, which is subject to change.
Parameters
IEnumerable<string> whitelist
A list of the string names of the properties uid should not include. If `None`, defaults to `_DEFAULT_UID_WHITE_LIST`, which includes most properties user allowes to change.
Returns
IEnumerator<string>
A uid string.

object uid_dyn(object whitelist)

Generates a 'Unique Identifier' based on all internal fields. (experimental)

Warning: THIS FUNCTION IS EXPERIMENTAL. It may change or be removed at any time, and without warning.

Caller should use the uid string to check `RunConfig` instance integrity in one session use, but should not rely on the implementation details, which is subject to change.
Parameters
object whitelist
A list of the string names of the properties uid should not include. If `None`, defaults to `_DEFAULT_UID_WHITE_LIST`, which includes most properties user allowes to change.
Returns
object
A uid string.

Public static methods

RunConfig NewDyn(object master, ImplicitContainer<T> num_cores, ImplicitContainer<T> log_device_placement, ImplicitContainer<T> gpu_memory_fraction, object tf_random_seed, ImplicitContainer<T> save_summary_steps, ImplicitContainer<T> save_checkpoints_secs, object save_checkpoints_steps, ImplicitContainer<T> keep_checkpoint_max, ImplicitContainer<T> keep_checkpoint_every_n_hours, ImplicitContainer<T> log_step_count_steps, object protocol, ImplicitContainer<T> evaluation_master, object model_dir, object session_config, ImplicitContainer<T> session_creation_timeout_secs)

Constructor. (deprecated)

Warning: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: When switching to tf.estimator.Estimator, use tf.estimator.RunConfig instead.

The superclass `ClusterConfig` may set properties like `cluster_spec`, `is_chief`, `master` (if `None` in the args), `num_ps_replicas`, `task_id`, and `task_type` based on the `TF_CONFIG` environment variable. See `ClusterConfig` for more details.

N.B.: If `save_checkpoints_steps` or `save_checkpoints_secs` is set, `keep_checkpoint_max` might need to be adjusted accordingly, especially in distributed training. For example, setting `save_checkpoints_secs` as 60 without adjusting `keep_checkpoint_max` (defaults to 5) leads to situation that checkpoint would be garbage collected after 5 minutes. In distributed training, the evaluation job starts asynchronously and might fail to load or find the checkpoint due to race condition.
Parameters
object master
TensorFlow master. Defaults to empty string for local.
ImplicitContainer<T> num_cores
Number of cores to be used. If 0, the system picks an appropriate number (default: 0).
ImplicitContainer<T> log_device_placement
Log the op placement to devices (default: False).
ImplicitContainer<T> gpu_memory_fraction
Fraction of GPU memory used by the process on each GPU uniformly on the same machine.
object tf_random_seed
Random seed for TensorFlow initializers. Setting this value allows consistency between reruns.
ImplicitContainer<T> save_summary_steps
Save summaries every this many steps.
ImplicitContainer<T> save_checkpoints_secs
Save checkpoints every this many seconds. Can not be specified with `save_checkpoints_steps`.
object save_checkpoints_steps
Save checkpoints every this many steps. Can not be specified with `save_checkpoints_secs`.
ImplicitContainer<T> keep_checkpoint_max
The maximum number of recent checkpoint files to keep. As new files are created, older files are deleted. If None or 0, all checkpoint files are kept. Defaults to 5 (that is, the 5 most recent checkpoint files are kept.)
ImplicitContainer<T> keep_checkpoint_every_n_hours
Number of hours between each checkpoint to be saved. The default value of 10,000 hours effectively disables the feature.
ImplicitContainer<T> log_step_count_steps
The frequency, in number of global steps, that the global step/sec will be logged during training.
object protocol
An optional argument which specifies the protocol used when starting server. None means default to grpc.
ImplicitContainer<T> evaluation_master
the master on which to perform evaluation.
object model_dir
directory where model parameters, graph etc are saved. If `None`, will use `model_dir` property in `TF_CONFIG` environment variable. If both are set, must have same value. If both are `None`, see `Estimator` about where the model will be saved.
object session_config
a ConfigProto used to set session parameters, or None. Note - using this argument, it is easy to provide settings which break otherwise perfectly good models. Use with care.
ImplicitContainer<T> session_creation_timeout_secs
Max time workers should wait for a session to become available (on initialization or when recovering a session) with MonitoredTrainingSession. Defaults to 7200 seconds, but users may want to set a lower value to detect problems with variable / session (re)-initialization more quickly.

Public properties

ClusterSpec cluster_spec get;

object cluster_spec_dyn get;

PythonFunctionContainer device_fn get;

object device_fn_dyn get;

object environment get;

object environment_dyn get;

object eval_distribute get;

object eval_distribute_dyn get;

string evaluation_master get;

object evaluation_master_dyn get;

object experimental_max_worker_delay_secs get;

object experimental_max_worker_delay_secs_dyn get;

Nullable<int> global_id_in_cluster get;

object global_id_in_cluster_dyn get;

bool is_chief get;

object is_chief_dyn get;

int keep_checkpoint_every_n_hours get;

object keep_checkpoint_every_n_hours_dyn get;

int keep_checkpoint_max get;

object keep_checkpoint_max_dyn get;

int log_step_count_steps get;

object log_step_count_steps_dyn get;

string master get;

object master_dyn get;

string model_dir get;

object model_dir_dyn get;

int num_ps_replicas get;

object num_ps_replicas_dyn get;

int num_worker_replicas get;

object num_worker_replicas_dyn get;

object protocol get;

object protocol_dyn get;

object PythonObject get;

Nullable<int> save_checkpoints_secs get;

object save_checkpoints_secs_dyn get;

Nullable<int> save_checkpoints_steps get;

object save_checkpoints_steps_dyn get;

int save_summary_steps get;

object save_summary_steps_dyn get;

object service get;

object service_dyn get;

object session_config get;

object session_config_dyn get;

int session_creation_timeout_secs get;

object session_creation_timeout_secs_dyn get;

int task_id get;

object task_id_dyn get;

object task_type get;

object task_type_dyn get;

object tf_config get;

object tf_config_dyn get;

Nullable<int> tf_random_seed get;

object tf_random_seed_dyn get;

object train_distribute get;

object train_distribute_dyn get;