You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@systemds.apache.org by "MaximilianSchreff (via GitHub)" <gi...@apache.org> on 2023/12/07 15:56:26 UTC
[PR] ResNet Bottleneck Architecture [systemds]
MaximilianSchreff opened a new pull request, #1957:
URL: https://github.com/apache/systemds/pull/1957
This series of commits adds the forward pass of the residual network bottleneck architecture as introduces by "Deep Residual Learning for Image Recognition" by Kaiming He et. al. and refined in "ResNet v1.5 for PyTorch" by NVIDIA.
The individual components (bottleneck residual block, bottleneck residual layer) has been unit tested and the tests were added to the NN test suite and the whole bottleneck ResNet has been tested against the implementation of PyTorch which was not added to the test suite.
@phaniarnab
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@systemds.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] ResNet Bottleneck Architecture [systemds]
Posted by "MaximilianSchreff (via GitHub)" <gi...@apache.org>.
MaximilianSchreff closed pull request #1957: ResNet Bottleneck Architecture
URL: https://github.com/apache/systemds/pull/1957
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@systemds.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] ResNet Bottleneck Architecture [systemds]
Posted by "MaximilianSchreff (via GitHub)" <gi...@apache.org>.
MaximilianSchreff commented on PR #1957:
URL: https://github.com/apache/systemds/pull/1957#issuecomment-2435502178
Content merged by other PR
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@systemds.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] ResNet Bottleneck Architecture [systemds]
Posted by "MaximilianSchreff (via GitHub)" <gi...@apache.org>.
MaximilianSchreff commented on PR #1957:
URL: https://github.com/apache/systemds/pull/1957#issuecomment-1872539593
@j143 Thank you for the feedback!
I'm still developing the backward pass for the models. I will create a fully training loop example and instructions once I implemented the backward pass.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@systemds.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] ResNet Bottleneck Architecture [systemds]
Posted by "MaximilianSchreff (via GitHub)" <gi...@apache.org>.
MaximilianSchreff commented on PR #1957:
URL: https://github.com/apache/systemds/pull/1957#issuecomment-1845606095
Notice that this neural network will likely only work in the training loop with the fixed bug of nested lists from #1956
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@systemds.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] ResNet Bottleneck Architecture [systemds]
Posted by "j143 (via GitHub)" <gi...@apache.org>.
j143 commented on PR #1957:
URL: https://github.com/apache/systemds/pull/1957#issuecomment-1859186991
Hi @MaximilianSchreff , thanks a lot for working on this. Nice to see that the code is document in the functions.
can we create one small .md file with instructions on how to run, limitations, a quick example! If you have any images/diagrams (even hand drawn rough diagrams also fine).
May be like this: https://pytorch.org/hub/nvidia_deeplearningexamples_resnet50/
Regards
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: dev-unsubscribe@systemds.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org