Alvaro Lopez Antelo

Results 9 comments of Alvaro Lopez Antelo

Timeouts are ocurring approximately every 45 minutes

Hi! Thank you for your detailed reply! I ran the modified code on my laptop, still connected to wifi but with a pretty stable connection and the errors are now...

Oops, I am still getting the same errors with no logging messages from the API in the try catch block. Will try to connect the Pi through ethernet

I have tested CUDA and pytorch using this demo code and it ran correctly on gpu https://github.com/jcjohnson/pytorch-examples pytorch tensors Also tested CUDNN with MNIST sample code from CUDNN All installation...

Solved the issue adding compute_30 to all setup.py and recompiling all torch modules. It was necesary for Tesla K80

> > @AlverGant hi bro, i try to add compute_30 like this > > ![image](https://user-images.githubusercontent.com/22334680/82791548-921b3a00-9e33-11ea-9957-387b4825fcf8.png) > > and recompiling modules but error persist. Hi @Exspiravit that is exactly what I...

You are having the same error that I had "cuda call failed" the differences is that I am using cuda 9 and torch 1.0.0 as recommended in the requirements from...

And I am using an older driver NVIDIA-SMI 384.130

[solved through ](https://github.com/eloquentarduino/tflm_esp32/issues/4)