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.
 
 
 
 
 
oahzxl 501a9e9cd2
[hotfix] polish flash attention (#1802)
2 years ago
..
amp
auto_parallel [autoparallel]add essential CommActions for broadcast oprands (#1793) 2 years ago
builder
cli
communication
context
device
engine
fx [autoparallel]add essential CommActions for broadcast oprands (#1793) 2 years ago
gemini [kernel] skip tests of flash_attn and triton when they are not available (#1798) 2 years ago
kernel [hotfix] polish flash attention (#1802) 2 years ago
logging
nn [hotfix] fix zero's incompatibility with checkpoint in torch-1.12 (#1786) 2 years ago
pipeline
registry
tensor [NFC] polish type hint for shape consistency (#1801) 2 years ago
testing
trainer
utils
zero [hotfix] fix zero's incompatibility with checkpoint in torch-1.12 (#1786) 2 years ago
__init__.py
constants.py
core.py
global_variables.py
initialize.py