Use TB protos where appropriate #1678
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
@nfelt I am not sure if I like the pattern of introducing two |
TensorFlow is no longer exposing Python binding for proto definition. TensorBoard has decided to copy all the required proto definitions over and use it where possible. We still rely on TensorFlow version of the proto for: - TB summary v1 APIs. e.g., *.summary.pb() - places where we use pywrap_tensorflow.EventsWriter This commit mostly made mechanical changes for anything that uses `tensor_util.make_tensor_proto` and anything that matches `tf\.(Event|Summary)`.
nfelt
approved these changes
Dec 12, 2018
tensorboard/plugins/custom_scalar/custom_scalars_plugin_test.py
Outdated
Show resolved
Hide resolved
stephanwlee
force-pushed
the
util_shim4
branch
from
9608769
to
fc829bd
Compare
stephanwlee
added a commit
to stephanwlee/tensorboard
that referenced
this pull request
Dec 22, 2018
Now that TensorBoard hosts all proto definitions required to run it, we use the TensorBoard version of protos. tensorflow#1678 has missed few usages of SessionLog.[CONSTANT] and this change fixes that.
stephanwlee
added a commit
that referenced
this pull request
Dec 26, 2018
stephanwlee
added a commit
to stephanwlee/tensorboard
that referenced
this pull request
Dec 26, 2018
We are deprecating usage of tf.proto but have missed few usage of RunMetadata in tensorflow#1678. Confirmed that this is not causing regression by: 1. running TB with variety of event files 2. reading the code: change in graphs_plugin is innocuous as we are serializing proto into string. Do note that this changes output of event_accumulators#RunMetadata but we do not seem to be using it at least in our repo.
stephanwlee
added a commit
that referenced
this pull request
Dec 27, 2018
We are deprecating usage of tf.proto but have missed few usage of RunMetadata in #1678. Confirmed that this is not causing regression by: 1. running TB with variety of event files 2. reading the code: change in graphs_plugin is innocuous as we are serializing proto into string. Do note that this changes output of event_accumulators#RunMetadata but we do not seem to be using it at least in our repo.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
TensorFlow is no longer exposing Python binding for proto definition.
TensorBoard has decided to copy all the required proto definitions over
and use it where possible.
We still rely on TensorFlow version of the proto for:
This commit mostly made mechanical changes for anything that
uses
tensor_util.make_tensor_protoand anything that matchestf\.(Event|Summary).