people-and-planet-ai.image-classification.e2e_test: test_predict failed

[flaky-bot]

This test failed!

To configure my behavior, see the Flaky Bot documentation.

If I'm commenting on this issue too often, add the flakybot: quiet label and
I will stop commenting.

---

commit: d249cad
buildURL: Build Status, Sponge
status: failed

Test output

Traceback (most recent call last):
  File "/workspace/people-and-planet-ai/image-classification/e2e_test.py", line 111, in model_endpoint_id
    deployed_model_id = deploy_model.deploy_model(
  File "/workspace/people-and-planet-ai/image-classification/deploy_model.py", line 80, in deploy_model
    deployed_model = response.result()
  File "/workspace/people-and-planet-ai/image-classification/.nox/py-3-8/lib/python3.8/site-packages/google/api_core/future/polling.py", line 134, in result
    raise self._exception
google.api_core.exceptions.InternalServerError: 500 INTERNAL

[dandhlee]

hey @bradmiro, this test failed on tfkeras_test.py, is this the same issue you were looking into earlier?

[leahecole]

@dandhlee this issue is actually for the people and planet ai (the data-science-onramp just also happened to fail, for that reason you mentioned 😄 ) - I'll reassign to David who owns the people and planet sample

[davidcavazos]

It looks like it failed while deploying the model.

failed on setup with "google.api_core.exceptions.InternalServerError: 500 INTERNAL"
Traceback (most recent call last):
  File "/workspace/people-and-planet-ai/image-classification/e2e_test.py", line 111, in model_endpoint_id
    deployed_model_id = deploy_model.deploy_model(
  File "/workspace/people-and-planet-ai/image-classification/deploy_model.py", line 80, in deploy_model
    deployed_model = response.result()
  File "/workspace/people-and-planet-ai/image-classification/.nox/py-3-8/lib/python3.8/site-packages/google/api_core/future/polling.py", line 134, in result
    raise self._exception
google.api_core.exceptions.InternalServerError: 500 INTERNAL

I haven't had that error before. Maybe it's an intermittent a backend issue?

[davidcavazos]

The tfkeras_test.py error looks more like it couldn't find the model file in GCS.

Traceback (most recent call last):
  File "/workspace/data-science-onramp/ai-platform/tfkeras_test.py", line 126, in test_tfkeras
    assert bucket.blob(f"{MODEL_DIR}/saved_model.pb").exists()
AssertionError: assert False
 +  where False = <bound method Blob.exists of <Blob: tfkeras-job-dir-c4882638-353e-483c-87ce-63037aa09af2, model/saved_model.pb, None>>()
 +    where <bound method Blob.exists of <Blob: tfkeras-job-dir-c4882638-353e-483c-87ce-63037aa09af2, model/saved_model.pb, None>> = <Blob: tfkeras-job-dir-c4882638-353e-483c-87ce-63037aa09af2, model/saved_model.pb, None>.exists
 +      where <Blob: tfkeras-job-dir-c4882638-353e-483c-87ce-63037aa09af2, model/saved_model.pb, None> = <bound method Bucket.blob of <Bucket: tfkeras-job-dir-c4882638-353e-483c-87ce-63037aa09af2>>('model/saved_model.pb')
 +        where <bound method Bucket.blob of <Bucket: tfkeras-job-dir-c4882638-353e-483c-87ce-63037aa09af2>> = <Bucket: tfkeras-job-dir-c4882638-353e-483c-87ce-63037aa09af2>.blob

[flaky-bot]

Test passed for commit 7ac9395 (Build Status, Sponge)! Closing this issue.