View source on GitHub |
Simple implementation of ClusterResolver that accepts all attributes.
Inherits From: ClusterResolver
tf.distribute.cluster_resolver.SimpleClusterResolver(
cluster_spec,
master='',
task_type=None,
task_id=None,
environment='',
num_accelerators=None,
rpc_layer=None
)
Please see the base class for documentation of arguments of its constructor.
It is useful if you want to specify some or all attributes.
Usage example with tf.distribute.Strategy
:
cluster = tf.train.ClusterSpec({"worker": ["worker0.example.com:2222",
"worker1.example.com:2222"]})
# On worker 0
cluster_resolver = SimpleClusterResolver(cluster, task_type="worker",
task_id=0,
num_accelerators={"GPU": 8},
rpc_layer="grpc")
strategy = tf.distribute.experimental.MultiWorkerMirroredStrategy(
cluster_resolver=cluster_resolver)
# On worker 1
cluster_resolver = SimpleClusterResolver(cluster, task_type="worker",
task_id=1,
num_accelerators={"GPU": 8},
rpc_layer="grpc")
strategy = tf.distribute.experimental.MultiWorkerMirroredStrategy(
cluster_resolver=cluster_resolver)
Attributes | |
---|---|
environment
|
Returns the current environment which TensorFlow is running in.
There are two possible return values, "google" (when TensorFlow is running in a Google-internal environment) or an empty string (when TensorFlow is running elsewhere). If you are implementing a ClusterResolver that works in both the Google environment and the open-source world (for instance, a TPU ClusterResolver or similar), you will have to return the appropriate string depending on the environment, which you will have to detect. Otherwise, if you are implementing a ClusterResolver that will only work in open-source TensorFlow, you do not need to implement this property. |
rpc_layer
|
|
task_id
|
Returns the task id this ClusterResolver indicates.
In TensorFlow distributed environment, each job may have an applicable task id, which is the index of the instance within its task type. This is useful when user needs to run specific code according to task index. For example,
Returns For more information, please see
|
task_type
|
Returns the task type this ClusterResolver indicates.
In TensorFlow distributed environment, each job may have an applicable task type. Valid task types in TensorFlow include 'chief': a worker that is designated with more responsibility, 'worker': a regular worker for training/evaluation, 'ps': a parameter server, or 'evaluator': an evaluator that evaluates the checkpoints for metrics. See Multi-worker configuration for more information about 'chief' and 'worker' task type, which are most commonly used. Having access to such information is useful when user needs to run specific code according to task types. For example,
Returns For more information, please see
|
Methods
cluster_spec
cluster_spec()
Returns the ClusterSpec passed into the constructor.
master
master(
task_type=None, task_id=None, rpc_layer=None
)
Returns the master address to use when creating a session.
Args | |
---|---|
task_type
|
(Optional) The type of the TensorFlow task of the master. |
task_id
|
(Optional) The index of the TensorFlow task of the master. |
rpc_layer
|
(Optional) The RPC used by distributed TensorFlow. |
Returns | |
---|---|
The name or URL of the session master. |
If a task_type and task_id is given, this will override the master
string passed into the initialization function.
num_accelerators
num_accelerators(
task_type=None, task_id=None, config_proto=None
)
Returns the number of accelerator cores per worker.
The SimpleClusterResolver does not do automatic detection of accelerators, and thus all arguments are unused and we simply return the value provided in the constructor.
Args | |
---|---|
task_type
|
Unused. |
task_id
|
Unused. |
config_proto
|
Unused. |