Omar Elayan 593de925cf generalize deepspeed linear and implement it for non cuda systems (#6932) ...

Co-authored-by: Olatunji Ruwase <olruwase@microsoft.com>
Co-authored-by: Logan Adams <114770087+loadams@users.noreply.github.com>

2025-01-29 17:35:11 +00:00

..

__init__.py

Rearrange inference OPS and stop using builder.load (#5490)

2024-10-09 01:22:28 +00:00

builder.py

Remove op compilation flags due to perf issue (#6944)

2025-01-13 16:50:22 +00:00

cpu_adam.py

CPUAdam fp16 and bf16 support (#5409)

2024-05-20 12:50:20 +00:00

fp_quantizer.py

generalize deepspeed linear and implement it for non cuda systems (#6932)

2025-01-29 17:35:11 +00:00

fused_adam.py

Fused adam for HPU (#5500)

2024-05-10 20:53:55 +00:00

no_impl.py

Add missing op_builder.hpu component for HPU accelerator (#4963)

2024-01-17 18:51:12 +00:00

transformer_inference.py

Rearrange inference OPS and stop using builder.load (#5490)

2024-10-09 01:22:28 +00:00