Samyamr/grad acc stage2 #338
Adding gradient accumulation support for ZeRO Stage 2. Changing all M…
a1cf86fd
Gradient Accumulation support for Stage 2. Model tests added to test …
0b12dc1e
Merge branch 'master' into samyamr/grad-acc-stage2
03d63caa
formatting
cafd9f2d
Merge branch 'samyamr/grad-acc-stage2' of github.com:microsoft/DeepSp…
d54ef562
Update deepspeed_light.py
1c88e9d0
Update ds_config_func_bs8_zero1.json
2f07fac9
tjruwase
approved these changes
on 2020-08-31
defining baseline prefix
67cf7280
Merge branch 'samyamr/grad-acc-stage2' of github.com:microsoft/DeepSp…
b7deede3
samyam
merged
7240abf3
into master 5 years ago
jeffra
deleted the samyamr/grad-acc-stage2 branch 4 years ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub