Closed papalotis closed 3 years ago
Yes, I had a similar issue. Could a commit be scheduled to remedy this problem? I attempted using a global change to the gpu device, but was unsuccessful in changing the device of a resident tensor in the EchoTorch library. I believe @papalotis' solution will work best for my ends.
Solved in 6159b46ed280df738b10e0aa3d290dc4314b883f (22/01/2021), dev branch No tests with NARMA-10 with CUDA
The ESN Algo breaks (throws an exception) when it is run with a GPU. More specifically the function
finalize
in/echotorch/nn/RRCell.py
.This error was created by running the Mackey Glass example The workaround that I have for now is
Definitely not the best solution but works for now