Connor Holmes
|
57e0a55066
Ensure is initialized for SD (#2534)
|
1 年之前 |
Jeff Rasley
|
2110552167
Fixes for torch 1.14 due to new torch.numel return type (#2522)
|
1 年之前 |
Connor Holmes
|
30c8d8a881
Initial dequant library implementation (#2521)
|
1 年之前 |
Jeff Rasley
|
0b26532612
Update codeowners (#2525)
|
1 年之前 |
Jeff Rasley
|
7ce371b139
Add note about nvcc/hipcc requirement (#2519)
|
1 年之前 |
Michael Wyatt
|
fe6785447d
Add missing Inference sub-configs (#2518)
|
1 年之前 |
Michael Wyatt
|
e59f80549e
Fix backward compatibility for InferenceConfig (#2516)
|
1 年之前 |
lokoppakmsft
|
78d4ca1f4b
Deepspeed quantization library v0.1 (#2450)
|
1 年之前 |
Lev Kurilenko
|
d40a15fcf0
Add max_tokens alias to max_out_tokens arg to maintain backwards compatibility (#2508)
|
1 年之前 |
Michael Wyatt
|
43bf035cfc
Update docs to autogenerate pydantic config model docs (#2509)
|
1 年之前 |
Ammar Ahmad Awan
|
b5d18a6ab3
DeepSpeed inference config. (#2459) (#2472)
|
1 年之前 |
Jeff Rasley
|
a4ceabb6b8
bump to 0.7.6
|
1 年之前 |
iLeGend
|
06e00f61ce
Fix typos: deepseed -> deepspeed (#2499)
|
1 年之前 |
lokoppakmsft
|
f2710bbe1d
Make data contiguous before the inplace reshape-copy_ function (#2489)
|
1 年之前 |
Michael Wyatt
|
be5ec506bd
Fix nightly CI tests (#2493)
|
1 年之前 |
Olatunji Ruwase
|
ee39187d8f
Make bf16_optimizer work for non pipeline (#2470)
|
1 年之前 |
Guo Yejun
|
3ca9878d8e
stage_1_and_2.py: no allreduce needed when mp size is 1 (#2494)
|
1 年之前 |
Connor Holmes
|
e7e7595502
Stable Diffusion Enhancements (#2491)
|
1 年之前 |
Kevin Ko
|
6f77da1bae
Add `scale_attn_by_inverse_layer_idx` feature (#2486)
|
1 年之前 |
Jeff Rasley
|
d2d1b4c3c7
[docs] add SD tutorial to deepspeed.ai news
|
1 年之前 |
Jeff Rasley
|
a63cb07edf
[docs] add SD tutorial to news
|
1 年之前 |
Michael Wyatt
|
521d329b97
Fix CI issues related to cupy install (#2483)
|
1 年之前 |
Reza Yazdani
|
9cfcf7431a
Add correct memory-allocation at DeepSpeed-Attention (#2474)
|
1 年之前 |
kyoto7250
|
a47c3e03cf
fix accelerate link (#2481)
|
1 年之前 |
savitamittal1
|
ffb6d98762
Added MLFLOW environment variables for logging metrics within trainig… (#2477)
|
1 年之前 |
Joe Mayer
|
4a06ecf631
Updating autotune json default in docs. (#2476)
|
1 年之前 |
Guo Yejun
|
f74ee31882
don't gather partitioned activations for mp size 1 (#2454)
|
1 年之前 |
Ammar Ahmad Awan
|
35458da0e0
Create a new folder structure to isolate model-specific code in DS (#2464)
|
1 年之前 |
Reza Yazdani
|
39bdc14195
fixing the checkpoint loading at inference-engine (#2429)
|
2 年之前 |
Connor Holmes
|
10e9d04c23
Cache Allocation and Softmax Fixes (#2433)
|
2 年之前 |