Add cupy and pynvrtc to GPU Dockerfile for custom networks in CUDA #403
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
guitarmind
changed the title
Contributor
|
Hi Mark, we recently updated our GPU image to use cuda 9.2. Closing this PR in favor of #416 another one with cupy for cuda 9.2. (I kept your commit). |
Contributor
Author
|
@rosbo Thanks a lot!!!! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Hi @rosbo ,
I'm participating Quora Insincere Questions Classification competition right now and it is limited to provide only 2-CPUs, 14GB RAM and a K80 GPU in 2 hours kernel run time. Therefore it is crucial for kagglers to be able to utilize CUDA library and its interfaces to build performant custom networks in Tensorflow or PyTorch under such limited resources.
cupy and pynvrtc are commonly required packages.
Some examples about their usage:
Quasi-Recurrent Neural Networks(QRNN)Independently Recurrent Neural Networks(IndRNN)I've built it successfully by following your instruction in the README.
That would be appreciated if you could help merge it!
Also feel free to let me know if any changes required to improve this PR, thanks!
Regards,
Mark