nv-inference.yml
Commit
1 year ago
- Files28
- .github/workflows
- nv-transformers-v100.yml
- setup-venv
- action.yml
- deepspeed
- comm
comm.py
- elasticity
- utils.py
- launcher
- runner.py
- runtime
- comm
- nccl.py
- fp16
- fused_optimizer.py
- onebit
- adam.py
- lamb.py
- zoadam.py
- unfused_optimizer.py
- utils.py
- docs/_tutorials
getting-started.md
- requirements
requirements-dev.txt- requirements-inf.txt
- requirements-triton.txt
- tests/unit
- checkpoint
- test_moe_checkpoint.py
- test_zero_optimizer.py
- compression
- test_compression.py
- model_parallelism
- test_configurable_parallel_mp.py
- test_configurable_parallel_pp.py
- moe
- test_moe.py
- test_moe_tp.py
- profiling/flops_profiler
- test_flops_profiler.py
- runtime
- half_precision
- onebit
- test_onebit.py
- test_fp16.py
- test_ds_initialize.py