tensorflow / hub

A library for transfer learning by reusing parts of TensorFlow models.
https://tensorflow.org/hub
Apache License 2.0
3.49k stars 1.67k forks source link

Bug: model weight bin file group1-shard1of2 is blocked in company network, which hitting the 403 issue. #893

Closed Stanven-1987 closed 1 year ago

Stanven-1987 commented 1 year ago

What happened?

Hi Team. I am trying to use the posenet within company network. However model weight bin file group1-shard1of2 is blocked in company network, which hitting the 403 issue. I have try to download the model files into the server, but rejected by company network. But the model.json can be retrieve. May I know if there is any other file type of the same model , let's say in JSON format? Appreciate you can advise. Thanks.

Relevant code

const model = await posenet.load({
                    architecture: 'MobileNetV1',
                    outputStride: 16,
                    multiplier: 0.75,
                    modelUrl: 'https://myapplicationserver.com/public/models/movenet/model.json'
                });

Relevant log output

In console: Failed to load resource: the server responded with status 403.
By direct click the model above company network saying:
The website you are attempting to access is blocked in line with company policy as it involves download a banned file type.

tensorflow_hub Version

0.12.0 (latest stable release)

TensorFlow Version

2.8 (latest stable release)

Other libraries

No response

Python Version

3.x

OS

Linux

singhniraj08 commented 1 year ago

@Stanven-1987,

Can you try downloading the model from TF-Hub site, save it in local file system and load the model as shown in example tutorial or by passing the local model file path in modelUrl param as shown here. Thanks.

Stanven-1987 commented 1 year ago

这是来自QQ邮箱的自动回复邮件。您好,您的邮件已收悉,谢谢。

Stanven-1987 commented 1 year ago

@Stanven-1987,

Can you try downloading the model from TF-Hub site, save it in local file system and load the model as shown in example tutorial or by passing the local model file path in modelUrl param as shown here. Thanks.

Thanks Niraj, I have try another way which solved it. I just simply rename the .bin extension, and modify the model.json to use the same file name. Then it works now.

Stanven-1987 commented 1 year ago

After simply rename the .bin extension, and modify the model.json to use the same file name. Then it works now.

google-ml-butler[bot] commented 1 year ago

Are you satisfied with the resolution of your issue? Yes No