tfdbg.add_debug_tensor_watch(
run_options,
node_name,
output_slot=0,
debug_ops='DebugIdentity',
debug_urls=None,
tolerate_debug_op_creation_failures=False,
global_step=-1
)
Defined in tensorflow/python/debug/lib/debug_utils.py
.
Add watch on a Tensor
to RunOptions
.
N.B.:
1. Under certain circumstances, the Tensor
may not get actually watched
(e.g., if the node of the Tensor
is constant-folded during runtime).
2. For debugging purposes, the parallel_iteration
attribute of all
tf.while_loop
s in the graph are set to 1 to prevent any node from
being executed multiple times concurrently. This change does not affect
subsequent non-debugged runs of the same tf.while_loop
s.
Args:
run_options
: An instance ofconfig_pb2.RunOptions
to be modified.node_name
: (str
) name of the node to watch.output_slot
: (int
) output slot index of the tensor from the watched node.debug_ops
: (str
orlist
ofstr
) name(s) of the debug op(s). Can be alist
ofstr
or a singlestr
. The latter case is equivalent to alist
ofstr
with only one element. For debug op types with customizable attributes, each debug op string can optionally contain a list of attribute names, in the syntax of: debug_op_name(attr_name_1=attr_value_1;attr_name_2=attr_value_2;...)debug_urls
: (str
orlist
ofstr
) URL(s) to send debug values to, e.g.,file:///tmp/tfdbg_dump_1
,grpc://localhost:12345
.tolerate_debug_op_creation_failures
: (bool
) Whether to tolerate debug op creation failures by not throwing exceptions.global_step
: (int
) Optional global_step count for this debug tensor watch.