|
About the Memory Format category
|
|
0
|
1588
|
March 18, 2020
|
|
Method for efficiently transferring non-autograd tensors to CPU from GPU?
|
|
4
|
34
|
April 9, 2025
|
|
Clip Grad Norm on GPU without sync
|
|
1
|
31
|
March 30, 2025
|
|
Redefining model with fewer parameters → out of memory (OOM)?
|
|
1
|
17
|
February 19, 2025
|
|
DtoH transfer of a partial tensor or untyped_storage to tensor without memory allocation
|
|
0
|
13
|
February 19, 2025
|
|
Questions about GPU memory usage
|
|
1
|
117
|
February 5, 2025
|
|
How can I decrease the pytorch confidence to hold too much reserved memory
|
|
3
|
93
|
January 8, 2025
|
|
`torch.cuda.is_available()` allocates unwanted memory?
|
|
2
|
491
|
December 11, 2024
|
|
Guarantee traversal order for optimiser states
|
|
0
|
82
|
October 12, 2024
|
|
How to share data among DataLoader processes to save memory
|
|
6
|
13719
|
October 10, 2024
|
|
Libtorch CPP Api for Memory Format Channels Last
|
|
1
|
39
|
September 7, 2024
|
|
Replacing torch.zeros internals with cudaMemset instead of fill kernel
|
|
2
|
177
|
September 5, 2024
|
|
Fancy idexing memory footprint
|
|
0
|
93
|
August 19, 2024
|
|
Unable to free all GPU memory
|
|
3
|
336
|
August 12, 2024
|
|
Why does tf.tile not make use of strided layout? And what about "inverse" strides?
|
|
0
|
68
|
August 3, 2024
|
|
Advanced Slicing
|
|
3
|
184
|
July 28, 2024
|
|
Why aren't inputs to conv1d channels last?
|
|
3
|
2119
|
July 22, 2024
|
|
Frombuffer() → "The given buffer is not writable"
|
|
1
|
899
|
June 19, 2024
|
|
TensorDataset with lazy loading?
|
|
4
|
1033
|
June 7, 2024
|
|
Understanding GPU memory visualization result
|
|
1
|
199
|
May 11, 2024
|
|
Do operations between tensors and scalars move the tensor to CPU?
|
|
5
|
1131
|
April 20, 2024
|
|
Understanding error msg "view size is not compatible with input tensor's size and stride"
|
|
5
|
3421
|
April 19, 2024
|
|
Torch + pytest leads to memory fragmentation: How to do proper integration testing of a lot of torch models?
|
|
0
|
350
|
April 19, 2024
|
|
Using 128 bit floating point datatype with Pytorch (not a complex number)
|
|
4
|
3807
|
March 31, 2024
|
|
What does the term "peak memory" should be referring to, max allocated vs. reserved?
|
|
0
|
245
|
March 16, 2024
|
|
Computing to a sub-tensor portion of the output tensor?
|
|
0
|
199
|
February 28, 2024
|
|
What is the most optimal shape of a tensor for storage and computational efficiency?
|
|
0
|
265
|
February 14, 2024
|
|
Large disk usage for some torch tensors (200MB vs 4MB) with same shape and dtype
|
|
2
|
394
|
January 22, 2024
|
|
Help with CUDA memory allocation during forward Linear
|
|
5
|
718
|
January 9, 2024
|
|
Discrepancy Between Expected and Actual GPU Memory Usage for Large Tensors
|
|
0
|
318
|
January 8, 2024
|