Dockerfile: Support CUDA 11 #45071
Dockerfile: Support CUDA 11 #45071
Conversation
💊 CI failures summary and remediationsAs of commit 6d3baa5 (more details on the Dr. CI page):
ci.pytorch.org: 1 failedThis comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions on the GitHub issue tracker or post in the (internal) Dr. CI Users group. This comment has been revised 1 time. |
Codecov Report
@@ Coverage Diff @@
## master #45071 +/- ##
==========================================
- Coverage 67.85% 67.85% -0.01%
==========================================
Files 384 384
Lines 50020 50020
==========================================
- Hits 33942 33941 -1
- Misses 16078 16079 +1
Continue to review full report at Codecov.
|
ailzhang
added
the
triaged
seemethere
left a comment
seemethere
left a comment
There was a problem hiding this comment.
looks mostly good to me, just one outstanding comment
| FROM conda as conda-installs | ||
| ARG INSTALL_CHANNEL=pytorch-nightly | ||
| RUN /opt/conda/bin/conda install -c "${INSTALL_CHANNEL}" -y pytorch torchvision cudatoolkit=10.1 && \ | ||
| RUN /opt/conda/bin/conda install -c "${INSTALL_CHANNEL}" -y pytorch torchvision cudatoolkit=11.0.221 && \ |
There was a problem hiding this comment.
Wouldn't this also need a -c nvidia or is the CUDA 11 cudatoolkit now available in the stock anaconda repository?
There was a problem hiding this comment.
Thanks for the review! This is the latest cudatoolkit available from Anaconda's repository. Thus, -c nvidia should not be necessary.
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@seemethere has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
|
@seemethere merged this pull request in 99242ec. |
7 participants
Although PyTorch already supports CUDA 11, the Dockerfile still relies on CUDA 10. This pull request upgrades all the necessary versions such that recent NVIDIA GPUs like A100 can be used.