|
Severe training slowdown after unfold-based patchification
|
|
1
|
6
|
February 8, 2026
|
|
Torch.export does not support specify partial output
|
|
2
|
13
|
February 8, 2026
|
|
Diskoffloading during reverse mode
|
|
0
|
7
|
February 8, 2026
|
|
Post training quantized model gets the error "Copying from quantized Tensor to non-quantized Tensor is not allowed" even though I'm not copying tensor
|
|
3
|
22
|
February 7, 2026
|
|
NaN Loss Issues with Precision 16 in PyTorch Lightning GAN Training
|
|
9
|
3926
|
February 6, 2026
|
|
[ROCm][CI] fp8 acceptable accuracy threshold
|
|
0
|
12
|
February 6, 2026
|
|
RNN memory management: RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation
|
|
2
|
18
|
February 6, 2026
|
|
Pretrained dendritic ResNet-18: 4x more parameter-efficient than ResNet-34
|
|
0
|
13
|
February 6, 2026
|
|
Pytorch not compatible with rtx 5050
|
|
5
|
56
|
February 6, 2026
|
|
Multi-GPU training with training loop that can skip backpropagation
|
|
1
|
125
|
February 6, 2026
|
|
At / tensor indexing helper for stable API
|
|
0
|
19
|
February 5, 2026
|
|
torch.compile(fullgraph=True) fails on nn.utils.rnn.pad_sequence with padding_side, while fullgraph=False succeeds
|
|
0
|
12
|
February 5, 2026
|
|
Implementing/Optimizing custom scatter_reduce op with 'memorized' indices
|
|
0
|
14
|
February 5, 2026
|
|
Can 'torch.profiler' support GB20X NVIDIA GPU architectures?
|
|
1
|
16
|
February 5, 2026
|
|
Feedback on resnet-18-perforated : resnet-34 accuracy on your use-case?
|
|
0
|
10
|
February 5, 2026
|
|
Onnx export using Torch gives wrong results
|
|
3
|
341
|
February 5, 2026
|
|
RuntimeError: Unexpected error from cudaGetDeviceCount()
|
|
16
|
83069
|
February 5, 2026
|
|
Efficient dataloadfer for sharded dataset
|
|
2
|
31
|
February 5, 2026
|
|
Fully_shard with 2D mesh (4,1) still runs all-gather / reduce-scatter on the shard dimension
|
|
0
|
8
|
February 5, 2026
|
|
Yolox continous issues without cuda
|
|
4
|
23
|
February 5, 2026
|
|
How Do I use Pytorch with RTX 5060 Ti
|
|
15
|
10176
|
February 5, 2026
|
|
Error: accessing tensor output of CUDAGraphs that has been overwritten by a subsequent run
|
|
1
|
692
|
February 5, 2026
|
|
Dictionary Input/Output in C++
|
|
2
|
37
|
February 4, 2026
|
|
Np.pi equivalent in Pytorch
|
|
4
|
40868
|
February 4, 2026
|
|
Is there the Pi greek (3.1415...) defined somewhere?
|
|
2
|
6440
|
February 4, 2026
|
|
Cannot import torchtext (libtorchtext.so: undefined symbol)
|
|
6
|
9549
|
February 4, 2026
|
|
Title: RTX 5080 (Blackwell) Training Stagnation vs RTX A4000 Normal Learning - Same Code, Different Outcomes
|
|
7
|
106
|
February 4, 2026
|
|
Extracting int8 weights and other quant params after convert_pt2e
|
|
0
|
17
|
February 3, 2026
|
|
Are we going to have linux support soon?
|
|
1
|
51
|
February 3, 2026
|
|
Is torch Muon optimizer compatible with FSDP/HSDP?
|
|
0
|
23
|
February 3, 2026
|