.. |
__init__.py
|
ef869377e9
DeepSpeed Data Efficiency Library (#2585)
|
1 年之前 |
all_ops.py
|
ffcf384673
Abstract accelerator (step 1) (#2504)
|
1 年之前 |
async_io.py
|
91e15593ea
Control ds_report output (#1622)
|
2 年之前 |
builder.py
|
b841628207
Drop Maxwell Support (#2574)
|
1 年之前 |
builder_names.py
|
ffcf384673
Abstract accelerator (step 1) (#2504)
|
1 年之前 |
cpu_adagrad.py
|
c3c8d5dd93
AMD support (#1430)
|
2 年之前 |
cpu_adam.py
|
737fee63ba
Update cpu_adam.py (#1968)
|
2 年之前 |
fused_adam.py
|
b85eb3b979
Fix build issues on Windows (#2428)
|
2 年之前 |
fused_lamb.py
|
b85eb3b979
Fix build issues on Windows (#2428)
|
2 年之前 |
quantizer.py
|
30c8d8a881
Initial dequant library implementation (#2521)
|
1 年之前 |
random_ltd.py
|
ef869377e9
DeepSpeed Data Efficiency Library (#2585)
|
1 年之前 |
sparse_attn.py
|
316c4a43e0
Add flake8 to pre-commit checks (#2051)
|
2 年之前 |
spatial_inference.py
|
e7e7595502
Stable Diffusion Enhancements (#2491)
|
1 年之前 |
stochastic_transformer.py
|
da1fe2f82c
Remove hard torch dependency at install (#1166)
|
3 年之前 |
transformer.py
|
b4fcd98ff0
Inference PP changes for neox (#1899)
|
2 年之前 |
transformer_inference.py
|
b841628207
Drop Maxwell Support (#2574)
|
1 年之前 |
utils.py
|
599258f979
ZeRO 3 Offload (#834)
|
3 年之前 |