Making large AI models cheaper, faster and more accessible
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
ver217 823f3b9cf4
[doc] add deepspeed citation and copyright (#2996)
2 years ago
..
checkpoint [hotfix] fix a running error in test_colo_checkpoint.py (#1387) 2 years ago
checkpoint_io [CheckpointIO] a uniform checkpoint I/O module (#1689) 2 years ago
data_sampler
model [doc] add deepspeed citation and copyright (#2996) 2 years ago
multi_tensor_apply [setup] support pre-build and jit-build of cuda kernels (#2374) 2 years ago
profiler [Gemini] clean no used MemTraceOp (#1970) 2 years ago
rank_recorder [pipeline/rank_recorder] fix bug when process data before backward | add a tool for multiple ranks debug (#1681) 2 years ago
tensor_detector [NFC] polish utils/tensor_detector/__init__.py code style (#1573) 2 years ago
__init__.py [ddp] add is_ddp_ignored (#2434) 2 years ago
activation_checkpoint.py [utils] Add use_reetrant=False in utils.activation_checkpoint (#1460) 2 years ago
checkpointing.py [utils] refactor parallel layers checkpoint and bcast model on loading checkpoint (#1548) 2 years ago
common.py Fix port exception type (#2925) 2 years ago
cuda.py [refactor] refactor the memory utils (#715) 3 years ago
memory.py [gemini] APIs to set cpu memory capacity (#809) 3 years ago
moe.py
timer.py