issues
search
pytorch
/
nestedtensor
[Prototype] Tools for the concurrent manipulation of variably sized Tensors.
BSD 3-Clause "New" or "Revised" License
252
stars
28
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Faster transpose - better warp shuffle
#422
cpuhrsch
closed
3 years ago
0
Faster batchnorm inference - less overhead, more blocks
#421
cpuhrsch
closed
3 years ago
0
Faster batchnorm inference
#420
cpuhrsch
closed
3 years ago
0
Faster padding - fuse fill, reduce overhead
#419
cpuhrsch
closed
3 years ago
0
Faster transpose
#418
cpuhrsch
closed
3 years ago
0
Conv2d fallback using padding and masking
#417
cpuhrsch
closed
3 years ago
0
Faster transpose_copy for conv2d_1x1
#416
cpuhrsch
closed
3 years ago
0
resnext101_32x4d oriented optimized implementations for packed memory.
#415
cpuhrsch
closed
3 years ago
0
Restrict nested dim to one - EfficientSizeNode
#414
cpuhrsch
closed
3 years ago
0
Restrict nested dim to one - tests and constructor
#413
cpuhrsch
closed
3 years ago
0
Classy vision model benchmark
#412
cpuhrsch
closed
3 years ago
0
Hi, how to check the shape of a nestedtensor in the pytorch?
#411
wangxiao5791509
closed
3 years ago
0
Bugged if statement
#410
hjalmarlucius
opened
3 years ago
0
Outstanding concept, difficult to start using
#409
hjalmarlucius
opened
3 years ago
8
Remove list layout
#408
cpuhrsch
closed
3 years ago
0
More efficient MHA - faster padding
#407
cpuhrsch
closed
3 years ago
0
More efficient MHA - recycle input_mask for sequence mask
#406
cpuhrsch
closed
3 years ago
0
More efficient MHA - use CPU numel and improve benchmark
#405
cpuhrsch
closed
3 years ago
0
More efficient MHA - packed input projection
#404
cpuhrsch
closed
3 years ago
0
Additional MHA test coverage
#403
cpuhrsch
closed
3 years ago
0
More efficient MHA - sequence_mask
#402
cpuhrsch
closed
3 years ago
0
More efficient MHA - opt_sizes
#401
cpuhrsch
closed
3 years ago
0
More efficient MHA - input mask generation
#400
cpuhrsch
closed
3 years ago
0
Switch some MHA case to custom CUDA kernels
#399
cpuhrsch
closed
3 years ago
0
Function to generate mask from NestedSize
#398
cpuhrsch
closed
3 years ago
0
Faster to_tensor_mask
#397
cpuhrsch
closed
3 years ago
0
More efficient embedding
#396
cpuhrsch
closed
3 years ago
0
Faster numel and add
#395
cpuhrsch
closed
3 years ago
0
CUDA kernel for padding
#394
cpuhrsch
closed
3 years ago
0
Move more masking code into C++
#393
cpuhrsch
closed
3 years ago
0
to_sparse_csr_tensor
#392
cpuhrsch
closed
3 years ago
0
More efficient add
#391
cpuhrsch
closed
3 years ago
0
Improve nn.Embedding performance
#390
cpuhrsch
closed
3 years ago
0
Flat gelu and shortcut dropout
#389
cpuhrsch
closed
3 years ago
0
Improve nn.Linear performance
#388
cpuhrsch
closed
3 years ago
0
Add floor_divide and pow.Scalar
#387
cpuhrsch
closed
3 years ago
0
Bind layernorm fastertransformer cuda kernel
#386
cpuhrsch
closed
3 years ago
0
Improve matmul performance
#385
cpuhrsch
closed
3 years ago
0
Update README.md
#384
cpuhrsch
closed
3 years ago
0
Update tutorial
#383
cpuhrsch
closed
3 years ago
0
Update tutorial
#382
cpuhrsch
closed
3 years ago
0
Set explicit torch dependency
#381
cpuhrsch
closed
3 years ago
0
Update README.md
#380
cpuhrsch
closed
3 years ago
0
Apply latest fbsync changes
#379
cpuhrsch
closed
3 years ago
0
Remove is_contiguous override to accommodate C10_DISABLE_TENSORIMPL_EXTENSIBILITY
#378
cpuhrsch
closed
3 years ago
0
Remove overrides to accomodate C10_DISABLE_TENSORIMPL_EXTENSIBILITY
#377
cpuhrsch
closed
3 years ago
0
More efficient SizeNode
#376
cpuhrsch
closed
3 years ago
0
Switch to using torch.inference_mode instead of static declaration
#375
cpuhrsch
closed
3 years ago
0
Bring back binary upload job
#374
cpuhrsch
closed
3 years ago
0
Remove environment image
#373
cpuhrsch
closed
3 years ago
0
Previous
Next