Deploying a tempo model onto a Docker runtime and immediately calling it will result in an error:
Traceback (most recent call last):
File "main.py", line 80, in <module>
main()
File "main.py", line 74, in main
pred = object_detector.predict(img)
File "/home/janis/.conda/envs/chrono/lib/python3.8/site-packages/tempo/serve/base.py", line 234, in predict
return self.remote_with_spec(model_spec, *args, **kwargs)
File "/home/janis/.conda/envs/chrono/lib/python3.8/site-packages/tempo/serve/base.py", line 269, in remote_with_spec
response_raw = requests.post(endpoint, json=req, headers=headers, verify=ingress_options.verify_ssl)
File "/home/janis/.conda/envs/chrono/lib/python3.8/site-packages/requests/api.py", line 119, in post
return request('post', url, data=data, json=json, **kwargs)
File "/home/janis/.conda/envs/chrono/lib/python3.8/site-packages/requests/api.py", line 61, in request
return session.request(method=method, url=url, **kwargs)
File "/home/janis/.conda/envs/chrono/lib/python3.8/site-packages/requests/sessions.py", line 530, in request
resp = self.send(prep, **send_kwargs)
File "/home/janis/.conda/envs/chrono/lib/python3.8/site-packages/requests/sessions.py", line 643, in send
r = adapter.send(request, **kwargs)
File "/home/janis/.conda/envs/chrono/lib/python3.8/site-packages/requests/adapters.py", line 498, in send
raise ConnectionError(err, request=request)
requests.exceptions.ConnectionError: ('Connection aborted.', ConnectionResetError(104, 'Connection reset by peer'))
python main.py data/good_images/TfL_20200315_1300_00001.01251.jpg 5.10s user 1.51s system 32% cpu 20.344 total
This is despite the fact that model.wait_ready() is called as part of the call to deploy. The issue seems to be with the docker Python library.
@cliveseldon is there a specific bug or reference issue that describes the behaviour?
Deploying a
tempo
model onto a Docker runtime and immediately calling it will result in an error:This is despite the fact that
model.wait_ready()
is called as part of the call todeploy
. The issue seems to be with thedocker
Python library.@cliveseldon is there a specific bug or reference issue that describes the behaviour?