Ted comments

Results 4 comments of

Ted

TorchServe no prediction when input data gets bigger (Backend worker did not respond in given time)

- The reason behind sending a list of input data instead of single input: The computational performance - I have increased responseTimeout, but it does not help

TorchServe no prediction when input data gets bigger (Backend worker did not respond in given time)

**@agunapal** Thank you so much for the response. For an easier communication, I have tried to simplify the custom handler while it repros the problem. For this purpose, I imagine...

TorchServe no prediction when input data gets bigger (Backend worker did not respond in given time)

@agunapal Thank you so much for the response. Your test shows that the input sample of 1000 works. By the way, there are some differences that I cannot get that...

TorchServe no prediction when input data gets bigger (Backend worker did not respond in given time)

Hi @agunapal, I have tested another transformer model with the same custom handler, there is no issue there. I think that the issue is an incompatibility between SimpleTransformer and PyTorch....