Hi,
I'm currently trying to get the DeepONet example working (ode.ipynb)
If I run the example, I get
'No supported gpu backend found!'
but if I remove this line:
os.environ["CUDA_VISIBLE_DEVICES"] = "3"
and update the constructor call of trainer by replacing gpus=-1 with:
accelerator='gpu', strategy="dp", devices=1,
I at least got it to Train.
It then crashes in this line:
out = model(grid_points).as_tensor.detach()[0]
with the Error:
RuntimeError: expand(torch.FloatTensor{[1, 30]}, size=[30]): the number of sizes provided (1) must be greater or equal to the number of dimensions in the tensor (2)
Can anyone give me advice on how to get this running?
Hi, I'm currently trying to get the DeepONet example working (ode.ipynb)
If I run the example, I get
'No supported gpu backend found!'
but if I remove this line:os.environ["CUDA_VISIBLE_DEVICES"] = "3"
and update the constructor call of
trainer
by replacinggpus=-1
with:accelerator='gpu', strategy="dp", devices=1,
I at least got it to Train.It then crashes in this line:
out = model(grid_points).as_tensor.detach()[0]
with the Error:RuntimeError: expand(torch.FloatTensor{[1, 30]}, size=[30]): the number of sizes provided (1) must be greater or equal to the number of dimensions in the tensor (2)
Can anyone give me advice on how to get this running?