|
Torch.compile + DDP Multi-Node: grad_norm becomes NaN starting from Epoch 2
|
|
0
|
5
|
December 19, 2025
|
|
Hello, I would like to ask about the merge process of PyTorch. Can it be merged after the PR is approved, and why does it show that the merge workflow is not scheduled?
|
|
2
|
16
|
December 19, 2025
|
|
Gpu devices: nvidia-smi and cuda.get_device_name() output appear inconsistent
|
|
9
|
11273
|
December 18, 2025
|
|
SDPA backends supporting attn_mask
|
|
1
|
7
|
December 18, 2025
|
|
When will sm120 support be available?
|
|
9
|
919
|
December 19, 2025
|
|
Request for confirmation on the official obsolescence / end-of-support dates
|
|
0
|
4
|
December 18, 2025
|
|
SOLVED: PyTorch 2.7.1+XPU Intel Arc Graphics Complete Setup Guide (Linux)"
|
|
13
|
1756
|
December 17, 2025
|
|
Intel A770 GPU + Debian13 -- XPU available: False
|
|
0
|
15
|
December 17, 2025
|
|
Using Pytorch profiler
|
|
0
|
12
|
December 17, 2025
|
|
Is it possible to use multi-cpu acceleration for tensor operations in Jupyter Notebook?
|
|
2
|
38
|
December 17, 2025
|
|
How to deal with pytorch GPU compatibility hell?
|
|
7
|
101
|
December 16, 2025
|
|
NaN Loss Issues with Precision 16 in PyTorch Lightning GAN Training
|
|
8
|
3704
|
December 16, 2025
|
|
Beginner assistance on architecture CONV1D
|
|
8
|
97
|
December 16, 2025
|
|
Multi-GPU for pre-trained model
|
|
0
|
12
|
December 16, 2025
|
|
Cannot Build from Source on Linux due to 'File name too long' error during Git clone
|
|
2
|
17
|
December 16, 2025
|
|
Mixed-type triangular solves
|
|
0
|
11
|
December 15, 2025
|
|
Performance collapse under high concurrency CPU inference (thread oversubscription)
|
|
0
|
13
|
December 15, 2025
|
|
Is there significance to the weights of the first / last layers?
|
|
1
|
46
|
December 14, 2025
|
|
XPU out of memory error with Intel Arc Graphics (Meteor Lake) despite sufficient system memory and reported XPU capacity
|
|
7
|
627
|
December 14, 2025
|
|
OSError: [WinError 182] The operating system cannot run %1
|
|
6
|
2517
|
December 13, 2025
|
|
GPU support and obsolescence
|
|
5
|
63
|
December 13, 2025
|
|
PyTorch install doesn't support cuda 12.6 sm_75
|
|
3
|
477
|
December 11, 2025
|
|
Should the bias in a Linear layer be considered when estimating FLOP?
|
|
4
|
42
|
December 11, 2025
|
|
How to properly run CUDA ops asynchronously across multiple streams in PyTorch?
|
|
1
|
49
|
December 10, 2025
|
|
Image Segmentation and object detection with Bounding Box
|
|
1
|
23
|
December 10, 2025
|
|
`torch.cuda.memory._record_memory_history()` does not work with env var PYTORCH_CUDA_ALLOC_CONF=backend:cudaMallocAsync
|
|
0
|
12
|
December 10, 2025
|
|
How to run the latest version of cuDNN with pytorch?
|
|
1
|
44
|
December 9, 2025
|
|
Pytorch hanging on backward() step on certain hardware
|
|
1
|
34
|
December 9, 2025
|
|
Thread safety between model.state_dict and optimizer.step()
|
|
0
|
15
|
December 9, 2025
|
|
Torch.onnx.export crash when converting ssdlite320_mobilenet_v3_large
|
|
1
|
25
|
December 9, 2025
|