EasyOCR Artifact

[gregd33]

Description

Same code to use it would be:

@bentoml.artifacts([
    EasyOCRArtifact("chinese_small")
])

and

service = EasyOCR()

lang_list = ['ch_sim', 'en']
recog_network = "zh_sim_g2"

model = easyocr.Reader(lang_list=lang_list, download_enabled=True, recog_network=recog_network )   
service.pack('chinese_small', model, lang_list=lang_list, recog_network= recog_network)

saved_path = service.save()

Artifact for EasyOCR

Motivation and Context

See #1538

How Has This Been Tested?

In progress

Types of changes

• Breaking change (fix or feature that would cause existing functionality to change)
• New feature and improvements (non-breaking change which adds/improves functionality)
• Bug fix (non-breaking change which fixes an issue)
• Code Refactoring (internal change which is not user facing)
• Documentation
• Test, CI, or build

Component(s) if applicable

• BentoService (service definition, dependency management, API input/output adapters)
• Model Artifact (model serialization, multi-framework support)
• Model Server (mico-batching, dockerisation, logging, OpenAPI, instruments)
• YataiService gRPC server (model registry, cloud deployment automation)
• YataiService web server (nodejs HTTP server and web UI)
• Internal (BentoML's own configuration, logging, utility, exception handling)
• BentoML CLI

Checklist:

To do

• My code follows the bentoml code style, both ./dev/format.sh and
  ./dev/lint.sh script have passed
  (instructions).
• My change reduces project test coverage and requires unit tests to be added
• I have added unit tests covering my code change
• My change requires a change to the documentation
• I have updated the documentation accordingly

[pep8speaks]

Hello @gregd33, Thanks for updating this PR.

There are currently no PEP 8 issues detected in this PR. Cheers! 🍻

Comment last updated at 2021-05-13 02:46:11 UTC

[codecov]

Codecov Report

Merging #1616 (6d22df3) into master (6f67144) will increase coverage by 0.11%.
The diff coverage is 87.71%.

@@            Coverage Diff             @@
##           master    #1616      +/-   ##
==========================================
+ Coverage   65.15%   65.26%   +0.11%     
==========================================
  Files         158      159       +1     
  Lines       11280    11337      +57     
==========================================
+ Hits         7349     7399      +50     
- Misses       3931     3938       +7     

Impacted Files	Coverage Δ
bentoml/artifact/__init__.py	0.00% <0.00%> (ø)
bentoml/frameworks/easyocr.py	89.28% <89.28%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 6f67144...6d22df3. Read the comment docs.

[gregd33]

One thing I'm uncertain about is if this is valid:

BentoML/bentoml/frameworks/easyocr.py

Line 53 in 9943276

def load(self, path, gpu=False):

Since the gpu will be determined on run time essentially (i.e. not when it is packed/saved) I thought it would make sense in load but am unsure how the user would use this arg.

Another question: I'm writing some tests now and need an image to send. Where can I store this image to access it for the test?

@yubozhao

[yubozhao]

One thing I'm uncertain about is if this is valid:

BentoML/bentoml/frameworks/easyocr.py

Line 53 in 9943276

def load(self, path, gpu=False):

Since the gpu will be determined on run time essentially (i.e. not when it is packed/saved) I thought it would make sense in load but am unsure how the user would use this arg.

Right now, it is difficult to modify the load behavior for model during loading BentoService. The current practice for setting additional option is in @artifacts decorator when define bento service. I think that's enough for most use-cases, where people would know ahead of using GPU or not in production.

Another question: I'm writing some tests now and need an image to send. Where can I store this image to access it for the test?

You can store a small test image in the same directory as the test files

[gregd33]

@yubozhao thanks. I've removed gpu and added the image. I think I have all the pieces in place now. I'll wait to see what the CI checks say

[gregd33]

Ready for review

[yubozhao]

@gregd33 It looks great. I think you will need to update the CI configuration and related files to make sure the integration test is running

You will need to update the CI, similar to the changes:

BentoML/.github/workflows/ci.yml

Lines 281 to 296 in f556850

	xgboost_integration_tests:
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v2
	with:
	fetch-depth: 0 # fetch all tags and branches
	- name: Setup python
	uses: actions/setup-python@v2
	with:
	python-version: 3.6.10
	- name: Install test dependencies
	run: ./ci/install_test_deps.sh.cmd
	- name: Run tests
	run: ./ci/xgboost_integration_tests.sh
	- name: Upload test coverage to Codecov
	uses: codecov/codecov-action@v1.0.12

You will need to add scripts that install and run the tests. Similar to these files:
https://github.com/bentoml/BentoML/blob/master/ci/xgboost_integration_tests.sh

Make sure to give the bash script correct permissions for running chmod +x

[gregd33]

Okay easyocr integration test now enabled and passing

[gregd33]

Realized that I omitted gpu completely instead of making it configurable at pack time. I've now fixed that.

[yubozhao]

Realized that I omitted gpu completely instead of making it configurable at pack time. I've now fixed that.

Did you push that change? I can't seem to find it

[gregd33]

Thought I did but guess not. Pushed now

[yubozhao]

@gregd33 Merged Thank you for contributing!