tf.saved_model.experimental.TrackableResource

View source on GitHub

Holds a Tensor which a tf.function can capture.

tf.saved_model.experimental.TrackableResource(
    device=''
)

A TrackableResource is most useful for stateful Tensors that require initialization, such as tf.lookup.StaticHashTable. TrackableResources are discovered by traversing the graph of object attributes, e.g. during tf.saved_model.save.

A TrackableResource has three methods to override:

• _create_resource should create the resource tensor handle.
• _initialize should initialize the resource held at self.resource_handle.
• _destroy_resource is called upon a TrackableResource's destruction and should decrement the resource's ref count. For most resources, this should be done with a call to tf.raw_ops.DestroyResourceOp.

Example usage:

class DemoResource(tf.saved_model.experimental.TrackableResource):
  def __init__(self):
    super().__init__()
    self._initialize()
  def _create_resource(self):
    return tf.raw_ops.VarHandleOp(dtype=tf.float32, shape=[2])
  def _initialize(self):
    tf.raw_ops.AssignVariableOp(
        resource=self.resource_handle, value=tf.ones([2]))
  def _destroy_resource(self):
    tf.raw_ops.DestroyResourceOp(resource=self.resource_handle)
class DemoModule(tf.Module):
  def __init__(self):
    self.resource = DemoResource()
  def increment(self, tensor):
    return tensor + tf.raw_ops.ReadVariableOp(
        resource=self.resource.resource_handle, dtype=tf.float32)
demo = DemoModule()
demo.increment([5, 1])
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([6., 2.], dtype=float32)>