Skip to content

Actions: microsoft/DeepSpeed

python

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
5,025 workflow runs
5,025 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

support autoTP with weight only quantization in DS inference path
python #11429: Pull request #4750 synchronize by ftian1
January 16, 2025 08:10 2m 32s ftian1:master
January 16, 2025 08:10 2m 32s
generalize deepspeed linear and implement it for non cuda systems
python #11426: Pull request #6932 synchronize by loadams
January 16, 2025 00:23 2m 10s oelayan7:linear
January 16, 2025 00:23 2m 10s
python
python #11425: Scheduled
January 16, 2025 00:09 2m 11s master
January 16, 2025 00:09 2m 11s
Pin numpy version
python #11424: Pull request #6953 opened by BLOrange-AMD
January 15, 2025 23:43 2m 11s ROCm:pin_numpy
January 15, 2025 23:43 2m 11s
python
python #11423: Merge group checks requested
January 15, 2025 22:09 2m 9s
January 15, 2025 22:09 2m 9s
Update sharded_moe.py to support top2 gate with Tutel
python #11422: Pull request #6948 synchronize by loadams
January 15, 2025 21:15 3m 4s xenshinu:patch-1
January 15, 2025 21:15 3m 4s
Update sharded_moe.py to support top2 gate with Tutel
python #11420: Pull request #6948 synchronize by xenshinu
January 15, 2025 19:40 Action required xenshinu:patch-1
January 15, 2025 19:40 Action required
python
python #11419: Merge group checks requested
January 15, 2025 19:25 2m 11s
January 15, 2025 19:25 2m 11s
warn to warning
python #11418: Pull request #6952 opened by qgallouedec
January 15, 2025 18:32 2m 13s qgallouedec:warn_to_warning
January 15, 2025 18:32 2m 13s
Addressing ipg Buffer Data Race Condition in Zero Stage2
python #11417: Pull request #3727 synchronize by loadams
January 15, 2025 17:09 Action required xxr3376:master
January 15, 2025 17:09 Action required
[inf] Add config var to enable keeping module on host
python #11416: Pull request #6846 synchronize by loadams
January 15, 2025 17:06 3m 24s oelayan7:keep_module_on_host
January 15, 2025 17:06 3m 24s
generalize deepspeed linear and implement it for non cuda systems
python #11415: Pull request #6932 synchronize by loadams
January 15, 2025 16:24 3m 58s oelayan7:linear
January 15, 2025 16:24 3m 58s
python
python #11414: Scheduled
January 15, 2025 00:09 3m 47s master
January 15, 2025 00:09 3m 47s
Set dataloader shuffle=true
python #11413: Pull request #6950 opened by loadams
January 14, 2025 23:52 2m 23s loadams/shuffle-true-dataloader
January 14, 2025 23:52 2m 23s
Update sharded_moe.py to support top2 gate with Tutel
python #11412: Pull request #6948 synchronize by xenshinu
January 14, 2025 20:11 4m 35s xenshinu:patch-1
January 14, 2025 20:11 4m 35s
Update sharded_moe.py to support top2 gate with Tutel
python #11411: Pull request #6948 opened by xenshinu
January 14, 2025 20:11 Action required xenshinu:patch-1
January 14, 2025 20:11 Action required
Bing/optimizer naming
python #11409: Pull request #3354 synchronize by loadams
January 14, 2025 17:03 2m 16s bing/optimizer-naming
January 14, 2025 17:03 2m 16s
python
python #11408: Scheduled
January 14, 2025 00:08 2m 11s master
January 14, 2025 00:08 2m 11s
Fix assert on Lamb optimizers with BF16
python #11407: Pull request #4451 synchronize by loadams
January 13, 2025 23:22 2m 8s loadams/lamb-bf16
January 13, 2025 23:22 2m 8s
Use ds-specific module id to avoid conflicts
python #11406: Pull request #6847 synchronize by loadams
January 13, 2025 22:41 2m 2s olruwase/pr_6772
January 13, 2025 22:41 2m 2s