Skip to content

Actions: deepspeedai/DeepSpeed

Actions

python

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
python.yml
2,224 workflow runs
2,224 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
python python #15601: Scheduled
1m 16s master
master
1m 16s
Add AutoEP + AutoTP parallel folding python #15600: Pull request #8064 synchronize by tohtana
1m 28s tohtana:tohtana/autoep-autotp-parallel-folding-design
tohtana:tohtana/autoep-autotp-parallel-folding-design
1m 28s
python python #15599: Merge group checks requested
1m 17s
1m 17s
Fix DeepCompile profile metadata backfill python #15598: Pull request #8094 synchronize by tohtana
1m 19s tohtana:tohtana/fix/deepcompile-device-time-schedule
tohtana:tohtana/fix/deepcompile-device-time-schedule
1m 19s
Fix DeepCompile profile metadata backfill python #15597: Pull request #8094 synchronize by tohtana
1m 25s tohtana:tohtana/fix/deepcompile-device-time-schedule
tohtana:tohtana/fix/deepcompile-device-time-schedule
1m 25s
Fix DeepCompile profile metadata backfill python #15596: Pull request #8094 opened by tohtana
1m 27s tohtana:tohtana/fix/deepcompile-device-time-schedule
tohtana:tohtana/fix/deepcompile-device-time-schedule
1m 27s
Add AutoEP + AutoTP parallel folding python #15595: Pull request #8064 synchronize by tohtana
1m 32s tohtana:tohtana/autoep-autotp-parallel-folding-design
tohtana:tohtana/autoep-autotp-parallel-folding-design
1m 32s
Add AutoEP + AutoTP parallel folding python #15594: Pull request #8064 synchronize by tohtana
1m 27s tohtana:tohtana/autoep-autotp-parallel-folding-design
tohtana:tohtana/autoep-autotp-parallel-folding-design
1m 27s
python python #15593: Merge group checks requested
1m 23s
1m 23s
Fix/warnings stacklevel mvapich runner python #15592: Pull request #7949 synchronize by sfc-gh-truwase
Action required nathon-lee:fix/warnings-stacklevel-mvapich-runner
nathon-lee:fix/warnings-stacklevel-mvapich-runner
Action required
Enable bf16 check_grad_overflow by default (matching fp16) python #15591: Pull request #8035 synchronize by yongzhe-wang
1m 25s yongzhe-wang:fix/bf16-check-grad-overflow-default-true
yongzhe-wang:fix/bf16-check-grad-overflow-default-true
1m 25s
python python #15590: Scheduled
1m 20s master
master
1m 20s
python python #15589: Merge group checks requested
1m 17s
1m 17s
python python #15588: Merge group checks requested
1m 14s
1m 14s
ZeRO 1/2: wait on all IPG-bucket producer streams in average_tensor (#8061) python #15587: Pull request #8080 synchronize by arunshar
Action required arunshar:fix/zero-overlap-comm-multistream
arunshar:fix/zero-overlap-comm-multistream
Action required
Warn when zero.Init silently falls back to a single rank (#8084) python #15586: Pull request #8089 synchronize by akshansh47
Action required akshansh47:fix/zero-init-unsharded-single-rank-warning
akshansh47:fix/zero-init-unsharded-single-rank-warning
Action required
ZeRO 1/2: wait on all IPG-bucket producer streams in average_tensor (#8061) python #15585: Pull request #8080 synchronize by arunshar
Action required arunshar:fix/zero-overlap-comm-multistream
arunshar:fix/zero-overlap-comm-multistream
Action required
Fix/warnings stacklevel mvapich runner python #15584: Pull request #7949 synchronize by nathon-lee
Action required nathon-lee:fix/warnings-stacklevel-mvapich-runner
nathon-lee:fix/warnings-stacklevel-mvapich-runner
Action required
Add numerical-correctness test for Muon under ZeRO-1/2 python #15583: Pull request #8091 opened by whycoming
Action required whycoming:test/muon-zero12-numerical-correctness
whycoming:test/muon-zero12-numerical-correctness
Action required
Reject Muon optimizer with reduce_scatter in ZeRO-1/2 python #15582: Pull request #8090 opened by whycoming
-1s whycoming:fix/muon-zero12-reduce-scatter-guard
whycoming:fix/muon-zero12-reduce-scatter-guard
-1s
Support AutoEP with ZeRO-3 zero.Init source modules python #15581: Pull request #8060 synchronize by delock
1m 27s tohtana:tohtana/autoep-zero3-zero-init
tohtana:tohtana/autoep-zero3-zero-init
1m 27s
Enable bf16 check_grad_overflow by default (matching fp16) python #15580: Pull request #8035 synchronize by yongzhe-wang
Action required yongzhe-wang:fix/bf16-check-grad-overflow-default-true
yongzhe-wang:fix/bf16-check-grad-overflow-default-true
Action required
python python #15579: Scheduled
1m 22s master
master
1m 22s
Warn when zero.Init silently falls back to a single rank (#8084) python #15578: Pull request #8089 opened by akshansh47
Action required akshansh47:fix/zero-init-unsharded-single-rank-warning
akshansh47:fix/zero-init-unsharded-single-rank-warning
Action required
fix: add buffer-length check in shm.cpp python #15577: Pull request #8082 synchronize by orbisai0security
Action required orbisai0security:fix-shm-parallel-memcpy-buffer-overflow
orbisai0security:fix-shm-parallel-memcpy-buffer-overflow
Action required