How to run evaluation code using pre-trained model locally?

[nitish11]

Hi,

I want to run evaluation code locally using pre-trained model on COCO-dataset. I don't want to create a docker container or any API based on call. I tried to run eval/eval.py by getting error related to azure setup. How to use existing code for "Generating images from the captions"?

[skabbit]

Got same question. If I run with my personal Azure Blob container, then I got azure.common.AzureMissingResourceHttpError: The specified container does not exist. ErrorCode: ContainerNotFound If I run with "attngan" container name, then azure.common.AzureException: HTTPSConnectionPool(host='attngan.blob.core.windows.net', port=443): Max retries exceeded with url: /images/2019/January/15/19_49_30_261376/bird_g0.png?comp=block&blockid=TURBd01EQXdNREF3TURBd01EQXdNREF3TURBd01EQXdNREF3TURBd01EQSUzRA%3D%3D (Caused by NewConnectionError('<urllib3.connection.VerifiedHTTPSConnection object at 0x1364e99e8>: Failed to establish a new connection: [Errno 8] nodename nor servname provided, or not known

I'm not familiar with Azure, but it seems, that container was removed or proper personal container must be deployed before starting Flask app.

[skabbit]

Removed all Azure Blob usage, replacing with saving to disk. All works well.

[lexical-kenobi]

Removed all Azure Blob usage, replacing with saving to disk. All works well.

can you put your repo link

[skabbit]

Sorry, has no repo for that. Just some brutal changes on the disk. I could check it, if you want, but it has been changed massively for my purposes and experiments.