Hello I am trying to convert a PyTorch EfficeintNet-B0 model to TFLite using ONNX. I am able to run the CPU version but the model will not bind to GPU. Is there any way to Bind the model to GPU? A method I tried of binding it to GPU does not yield the right output and the model returns NaNs.
Hi @atharva001,
Sorry for the delayed response. I think you might have solved the issue by this time. However, could you please make a try to resolve your issue with new AI-Edge-Torch, you can find more information here: googleblog. Please try them out and let us know if this resolves your issue. If you still need further help, feel free to open a new issue at the respective repos.
Thank You