implement tensorflow tensor handlers #421
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
|
Hello @hrmthw, Thanks for updating this PR. There are currently no PEP 8 issues detected in this PR. Cheers! 🍻 Comment last updated at 2019-12-19 02:42:48 UTC |
bojiang
changed the title
parano
reviewed
Dec 14, 2019
| """ | ||
| output_format = request.headers.get("output", "json") | ||
| if output_format not in {"json", "str"}: | ||
| return make_response( |
There was a problem hiding this comment.
use raise BadInput instead? we recently made some changes to how we handle this type of error. Similarily below in L157
parano
reviewed
Dec 14, 2019
| try: | ||
| import tensorflow as tf | ||
| except ImportError: | ||
| raise ImportError( |
There was a problem hiding this comment.
use raise MissingDependencyException instead? you may need to rebase or merge master
|
looks like you still need to add TensorFlow to the test dependencies, tests are still failing on travis-ci due to this. Consider mocking tf to avoid actually adding the dependency |
bojiang
changed the title
Codecov Report
@@ Coverage Diff @@
## master #421 +/- ##
==========================================
+ Coverage 58.55% 58.56% +<.01%
==========================================
Files 82 83 +1
Lines 5569 5705 +136
==========================================
+ Hits 3261 3341 +80
- Misses 2308 2364 +56
Continue to review full report at Codecov.
|
bojiang
changed the title
joshuacwnewton
added a commit
to MLH-Fellowship/BentoML
that referenced
this pull request
Jul 24, 2020
Model service works before saving, but saving/loading TF1 models is somewhat tricky. BentoML throws a warning about unsupported behavior. See bentoml#421 for more information.
parano
pushed a commit
that referenced
this pull request
Jul 31, 2020
* Rename Tensorflow tests to Tensorflow_v2.2 * Add partially working TF1.14 integration test Model service works before saving, but saving/loading TF1 models is somewhat tricky. BentoML throws a warning about unsupported behavior. See #421 for more information. * Refactor TF2 artifact tests using fixtures * Write test for TF2/Docker integration * Change scope of tf2_svc_saved_dir fixture * Fix bug for TF2.2 docker test (missing async) * Fix incorrectly formatted data in API request * Replace tmpdir with tmp_path_factory for scoping tmpdir can't be used with module scope in a pytest fixture. * Rename tf1 test objects to use '1' signifier * Replace '__call__' with 'call' to fix bug * Add prediction to build model layers before saving Solves issue where model could not serialized properly using .save(). * Fix request in TF2 docker test Return format and input data format were incorrect for TF API. * Use common test data across all TF2 tests * Remove partially working TF1.14 test * Run dev/format.sh and dev/lint.sh * Rename TF bento_service_example to TF2 This is because there will be another bento service example for TF1. * Remove _wait_until_ready and import instead * Add notes about image/host funcs from conftest.py
aarnphm
pushed a commit
to aarnphm/BentoML
that referenced
this pull request
Jul 29, 2022
* init tf_tensor_handler test * implement tensorflow tensor http_handler * implement tensorflow tensor cli_handler and lambda_handler * move input tensor auto transforming into artifact * fix tf_tensor_handler test * add new arg 'method' to TensorflowTensorHandler * add NestedDecoder; decode tensor handler output * add warning about importing tf 1.x SavedModel * style
aarnphm
pushed a commit
to aarnphm/BentoML
that referenced
this pull request
Jul 29, 2022
…ml#937) * Rename Tensorflow tests to Tensorflow_v2.2 * Add partially working TF1.14 integration test Model service works before saving, but saving/loading TF1 models is somewhat tricky. BentoML throws a warning about unsupported behavior. See bentoml#421 for more information. * Refactor TF2 artifact tests using fixtures * Write test for TF2/Docker integration * Change scope of tf2_svc_saved_dir fixture * Fix bug for TF2.2 docker test (missing async) * Fix incorrectly formatted data in API request * Replace tmpdir with tmp_path_factory for scoping tmpdir can't be used with module scope in a pytest fixture. * Rename tf1 test objects to use '1' signifier * Replace '__call__' with 'call' to fix bug * Add prediction to build model layers before saving Solves issue where model could not serialized properly using .save(). * Fix request in TF2 docker test Return format and input data format were incorrect for TF API. * Use common test data across all TF2 tests * Remove partially working TF1.14 test * Run dev/format.sh and dev/lint.sh * Rename TF bento_service_example to TF2 This is because there will be another bento service example for TF1. * Remove _wait_until_ready and import instead * Add notes about image/host funcs from conftest.py
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.
(Thanks for sending a pull request! Please make sure to read the contribution guidelines, then fill out the blanks below.)
What changes were proposed in this pull request?
implement tensorflow tensor handlers(auto-transform in artifact)
Does this close any currently open issues?
no
How was this patch tested?
57% on tensorflow 2, 1.14