build: Dependency bump - #1037
Conversation
Signed-off-by: oliver könig <okoenig@nvidia.com>
0e33251 to
0527fc1
Compare
|
/ok to test b83c274 |
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
…endency-bump Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
Signed-off-by: Charlie Truong <chtruong@nvidia.com>
| SLACK_WEBHOOK: ${{ secrets.SLACK_RELEASE_ENDPOINT }} | ||
| SLACK_WEBHOOK_ADMIN: ${{ secrets.SLACK_WEBHOOK_ADMIN }} | ||
| GH_TOKEN: ${{ secrets.PAT }} | ||
| # build-test-publish-wheel: |
There was a problem hiding this comment.
Disabling this for now because the targeted MCore and TE versions are not pip installable
| timeout: 40 | ||
| - script: L2_Launch_converter | ||
| - script: L2_Launch_models | ||
| timeout: 40 |
There was a problem hiding this comment.
Had to extend this a bit to get them passing. I think focus for now should be for validating functionality especially with the dependency bump.
| with: | ||
| package-name: megatron.bridge | ||
| python-binary: ${{ env.UV_PROJECT_ENVIRONMENT }}/bin/python | ||
| # pip-test-pytorch: |
There was a problem hiding this comment.
Disabling this for now because the targeted MCore and TE versions are not pip installable
| megatron-core = { path = "3rdparty/Megatron-LM/" } | ||
| mamba-ssm = { git = "https://github.com/state-spaces/mamba.git", rev = "6b32be06d026e170b3fdaf3ae6282c5a6ff57b06" } | ||
| nvidia-resiliency-ext = { git = "https://github.com/NVIDIA/nvidia-resiliency-ext.git", rev = "54f85fe422d296cf04ea524130014bd3a2c3add1" } | ||
| causal-conv1d = { git = "https://github.com/Dao-AILab/causal-conv1d.git", rev = "9d700d167c4ad299b0a5265ed1bdb4ee4a0ca111" } |
There was a problem hiding this comment.
These have to be built from source to make it work it seems. Unless there's cuda 13 builds for them we can reference.
|
|
||
| CUDA_VISIBLE_DEVICES="0,1" coverage run -a --data-file=/opt/Megatron-Bridge/.coverage --source=/opt/Megatron-Bridge/ -m pytest \ | ||
| --timeout=0.5 \ | ||
| --timeout=0.75 \ |
There was a problem hiding this comment.
Needed to increase this to get the unit tests working. It seemed that different tests would fail at times depending on the run. So didn't want to chase after those.
No description provided.