tfdbg.WatchOptions
tfdbg.WatchOptions
class tfdbg.WatchOptions
Defined in tensorflow/python/debug/wrappers/framework.py
.
Type for return values of watch_fn.
Methods
__init__
__init__( debug_ops=None, node_name_regex_whitelist=None, op_type_regex_whitelist=None, tensor_dtype_regex_whitelist=None, tolerate_debug_op_creation_failures=False )
Constructor of WatchOptions: Debug watch options.
Used as return values of watch_fn
s.
Args:
-
debug_ops
: (str
orlist of str
) Debug ops to be used. -
node_name_regex_whitelist
: Regular-expression whitelist for node_name, e.g.,"(weight_[0-9]+|bias_.*)"
-
op_type_regex_whitelist
: Regular-expression whitelist for the op type of nodes, e.g.,"(Variable|Add)"
. If bothnode_name_regex_whitelist
andop_type_regex_whitelist
are set, the two filtering operations will occur in a logicalAND
relation. In other words, a node will be included if and only if it hits both whitelists. -
tensor_dtype_regex_whitelist
: Regular-experssion whitelist for Tensor data type, e.g.,"^int.*"
. This whitelist operates in logicalAND
relations to the two whitelists above. -
tolerate_debug_op_creation_failures
: (bool
) whether debug op creation failures (e.g., due to dtype incompatibility) are to be tolerated by not throwing exceptions.
© 2017 The TensorFlow Authors. All rights reserved.
Licensed under the Creative Commons Attribution License 3.0.
Code samples licensed under the Apache 2.0 License.
https://www.tensorflow.org/api_docs/python/tfdbg/WatchOptions