Commits · 44bd538b110ce0e8fc69626854631c3aee0dc094 · OpenDAS / deepspeed

23 Dec, 2020 1 commit
- Elastic training support (#602) · 81aeea36
  Jeff Rasley authored Dec 22, 2020
```
Co-authored-by: Samyam Rajbhandari <samyamr@microsoft.com>
```
  81aeea36
11 Dec, 2020 1 commit
- [build] fix computer capability arch flags, add PTX, handle PTX (#591) · 8a184b6b
  Stas Bekman authored Dec 11, 2020
```
* fix arch flags, add PTX

* bug fix
Co-authored-by: Jeff Rasley <jerasley@microsoft.com>
```
  8a184b6b
07 Dec, 2020 2 commits

[build] add compute_86 (#577) · e8b126d9

Stas Bekman authored Dec 07, 2020

RTX-30 series are compute_86
```
python -c "import torch; print(torch.cuda.get_device_capability())"
```
This PR adds support for this compute capability.

Reference: https://developer.nvidia.com/cuda-gpus

Co-authored-by: Jeff Rasley <jerasley@microsoft.com>

e8b126d9

[build] make builder smarter and configurable wrt compute capabilities + docs (#578) · ce363d0e
Stas Bekman authored Dec 07, 2020

ce363d0e

03 Dec, 2020 2 commits
- [build] build against installed cuda-11.1 while torch built w/ cuda-11.0 (#570) · ff58fa7e
  Stas Bekman authored Dec 02, 2020
  
  ff58fa7e
- Add compute capability 8.0 if on cuda 11+ (#572) · be33bea4
  Jeff Rasley authored Dec 02, 2020
  
  be33bea4
12 Nov, 2020 1 commit

DeepSpeed JIT op + PyPI support (#496) · 31f46fee

Jeff Rasley authored Nov 12, 2020


Co-authored-by: Shaden Smith <Shaden.Smith@microsoft.com>
Co-authored-by: Reza Yazdani <reyazda@microsoft.com>

31f46fee