.. |
adagrad
|
69af361167
CPUAdam fp16 and bf16 support (#5409)
|
5 月之前 |
adam
|
69af361167
CPUAdam fp16 and bf16 support (#5409)
|
5 月之前 |
aio
|
6eefc3d0ea
Fix Memory Leak In AIO (#6630)
|
4 天之前 |
cpu
|
ec6cbb3c08
[CPU] Allow deepspeed.comm.inference_all_reduce in torch.compile graph (#5604)
|
3 月之前 |
deepspeed4science
|
297a6840e1
Update clang-format version from 16 to 18. (#5839)
|
2 月之前 |
fp_quantizer
|
c27483933d
wrap include cuda_bf16.h with ifdef BF16_AVAILABLE (#6520)
|
1 月之前 |
gds
|
6eefc3d0ea
Fix Memory Leak In AIO (#6630)
|
4 天之前 |
includes
|
297a6840e1
Update clang-format version from 16 to 18. (#5839)
|
2 月之前 |
lamb
|
c7724c6181
Switch from HIP_PLATFORM_HCC to HIP_PLATFORM_AMD (#4539)
|
1 年之前 |
lion
|
69af361167
CPUAdam fp16 and bf16 support (#5409)
|
5 月之前 |
quantization
|
e7dd28a23d
Fixed the Windows build. (#5596)
|
4 月之前 |
random_ltd
|
76c9c69fb1
Rocm warp size fix (#5402)
|
5 月之前 |
sparse_attention
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 年之前 |
spatial
|
c7724c6181
Switch from HIP_PLATFORM_HCC to HIP_PLATFORM_AMD (#4539)
|
1 年之前 |
transformer
|
645639bcf8
Rearrange inference OPS and stop using builder.load (#5490)
|
1 周之前 |
utils
|
b361c72761
Update DeepSpeed copyright license to Apache 2.0 (#3111)
|
1 年之前 |
xpu
|
d45cfd3455
[XPU] Support DeepNVMe new code structure (#6532)
|
3 周之前 |