|
About the Memory Format category
|
|
0
|
1643
|
March 18, 2020
|
|
Cuda Memory Use Batched Matrix Multiplication
|
|
7
|
2131
|
February 11, 2026
|
|
Sparse bmm cause CUDA misaligned address error
|
|
2
|
131
|
October 23, 2025
|
|
Free all GPU memory used in between runs
|
|
4
|
29767
|
August 18, 2025
|
|
Output mismatch with channels_last_3d?
|
|
1
|
124
|
July 9, 2025
|
|
Question: What takes so much gpu memory?
|
|
2
|
165
|
July 6, 2025
|
|
Understand the memory allocation visualization
|
|
1
|
180
|
July 1, 2025
|
|
Why the reserved memory is much larger than occupied memory?
|
|
1
|
164
|
June 25, 2025
|
|
Why does GPU memory usage not double when loading two identical models in PyTorch?
|
|
2
|
157
|
June 9, 2025
|
|
Questions about GPU memory usage
|
|
2
|
442
|
May 8, 2025
|
|
Out of Memory During LoRA Fine-Tuning on LLAMA-4-Scout-17B with H100 (80GB VRAM)
|
|
0
|
298
|
May 4, 2025
|
|
Method for efficiently transferring non-autograd tensors to CPU from GPU?
|
|
4
|
195
|
April 9, 2025
|
|
Clip Grad Norm on GPU without sync
|
|
1
|
156
|
March 30, 2025
|
|
Redefining model with fewer parameters → out of memory (OOM)?
|
|
1
|
103
|
February 19, 2025
|
|
DtoH transfer of a partial tensor or untyped_storage to tensor without memory allocation
|
|
0
|
84
|
February 19, 2025
|
|
How can I decrease the pytorch confidence to hold too much reserved memory
|
|
3
|
576
|
January 8, 2025
|
|
`torch.cuda.is_available()` allocates unwanted memory?
|
|
2
|
630
|
December 11, 2024
|
|
Guarantee traversal order for optimiser states
|
|
0
|
151
|
October 12, 2024
|
|
How to share data among DataLoader processes to save memory
|
|
6
|
14732
|
October 10, 2024
|
|
Libtorch CPP Api for Memory Format Channels Last
|
|
1
|
140
|
September 7, 2024
|
|
Replacing torch.zeros internals with cudaMemset instead of fill kernel
|
|
2
|
473
|
September 5, 2024
|
|
Fancy idexing memory footprint
|
|
0
|
189
|
August 19, 2024
|
|
Unable to free all GPU memory
|
|
3
|
590
|
August 12, 2024
|
|
Why does tf.tile not make use of strided layout? And what about "inverse" strides?
|
|
0
|
153
|
August 3, 2024
|
|
Advanced Slicing
|
|
3
|
318
|
July 28, 2024
|
|
Why aren't inputs to conv1d channels last?
|
|
3
|
2394
|
July 22, 2024
|
|
Frombuffer() → "The given buffer is not writable"
|
|
1
|
1905
|
June 19, 2024
|
|
TensorDataset with lazy loading?
|
|
4
|
1981
|
June 7, 2024
|
|
Understanding GPU memory visualization result
|
|
1
|
263
|
May 11, 2024
|
|
Do operations between tensors and scalars move the tensor to CPU?
|
|
5
|
1395
|
April 20, 2024
|