-
Encountered when using DDP. How should I locate the warning at this location?
Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
/home/ps/anaconda3/en…
-
I have a setup with `torch.Lightning` where I'm using custom `torchmetrics.Metric` as loss function contributions. Now I want to be able to do it with `ddp` by setting `dist_sync_on_step=True`, but th…
-
In the discussion in https://github.com/pytorch/pytorch/pull/53898#issuecomment-797760860 it was found that many advanced indexing operations have branches for large tensors by using `cuda::detail::ca…
-
Hi all,
I modified the hello_world model to perform a single MatMul operation instead of the Conv2d/Relu operations, and i'm unable to make it run on the NPU.
The code is mostly the same, the ch…
-
Hello!
I'm currently checking flash attention v2 and noticed that when copying from global memory to shared memory, the entire HeadDim (the K dimension in MNK tiling) needs to be copied to shared m…
-
The network is optimized using O2 optimization on -->
Cuda 10.0
PyTorch 1.0.0
CudNN 7.6.3
All conv2d operations have input output channels which are a multiple of 8. I did profiling of the code u…
-
Hi,
Facing this issue when use tensorrt 7.1.3 + pytorch 1.7.0 + torchvision 0.8.1
The network is a very simple MLP.
```
Warning: Encountered known unsupported method torch.Tensor.add
Warning…
-
I did everything in the README.
(I am using CPU only)
When running python app.py I get: AssertionError: Torch not compiled with CUDA enabled
-
### Describe the bug
I run the training but get this error
### Reproduction
Run accelerate config
```
compute_environment: LOCAL_MACHINE
debug: false
distributed_type: FSDP
downcast_bf16: 'n…
-
Thanks for your great work!!
After run "pip install ." , successfully installed VHAP-0.0.1 nvdiffrast-0.3.1
But I cannot run track.py as no nvdiffrast_plugin_gl.so error. Any suggestion?
The…