Corrdiff - Make IO Asynchronous with inference #305
Merged
Conversation
This file contains 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
mnabian
approved these changes
Jan 18, 2024
|
/blossom-ci |
MortezaMardani
approved these changes
Jan 18, 2024
|
/blossom-ci |
|
/blossom-ci |
NickGeneva
pushed a commit
to NickGeneva/modulus
that referenced
this pull request
Jan 26, 2024
* Refactor saving results to improve perf and overlap with inferencing * Update blossom-ci.yml (NVIDIA#295) * Change pip install commands with the correct PyPI package name (NVIDIA#298) * Make number of writer workers parameterized * add comment for writer workers * revert readme and blossom-ci * 2nd revert of readme * fix black formatting --------- Co-authored-by: David Pruitt <dpruitt@login-eos01.eos.clusters.nvidia.com> Co-authored-by: Kaustubh Tangsali <71059996+ktangsali@users.noreply.github.com> Co-authored-by: Abdullah <37012364+Saydemr@users.noreply.github.com>
NickGeneva
pushed a commit
that referenced
this pull request
Jan 26, 2024
* Refactor saving results to improve perf and overlap with inferencing * Update blossom-ci.yml (#295) * Change pip install commands with the correct PyPI package name (#298) * Make number of writer workers parameterized * add comment for writer workers * revert readme and blossom-ci * 2nd revert of readme * fix black formatting --------- Co-authored-by: David Pruitt <dpruitt@login-eos01.eos.clusters.nvidia.com> Co-authored-by: Kaustubh Tangsali <71059996+ktangsali@users.noreply.github.com> Co-authored-by: Abdullah <37012364+Saydemr@users.noreply.github.com>
mnabian
added a commit
that referenced
this pull request
Feb 21, 2024
* Fix TypeError in CorrDiff dataloader (#290) * Update dataset.py * Update formatting --------- Co-authored-by: Mohammad Amin Nabian <m.a.nabiyan@gmail.com> Co-authored-by: Mohammad Amin Nabian <mnabian@nvidia.com> * Update to 0.5.0 version (#293) * Fix diffusion example imports. (#294) * Fix pydantic issues (#302) * Fix minor bugs in corrdiff training (#303) * fix small bugs * formatting * add documentation for task * Fix zenith angle import, pydantic issues (#304) * Corrdiff - Make IO Asynchronous with inference (#305) * Refactor saving results to improve perf and overlap with inferencing * Update blossom-ci.yml (#295) * Change pip install commands with the correct PyPI package name (#298) * Make number of writer workers parameterized * add comment for writer workers * revert readme and blossom-ci * 2nd revert of readme * fix black formatting --------- Co-authored-by: David Pruitt <dpruitt@login-eos01.eos.clusters.nvidia.com> Co-authored-by: Kaustubh Tangsali <71059996+ktangsali@users.noreply.github.com> Co-authored-by: Abdullah <37012364+Saydemr@users.noreply.github.com> * CorrDiff integration: separate regression and diffusion configs (#308) * Update blossom-ci.yml (#295) * Change pip install commands with the correct PyPI package name (#298) * update configs * remove mixture loss * remove mixture loss from init --------- Co-authored-by: Kaustubh Tangsali <71059996+ktangsali@users.noreply.github.com> Co-authored-by: Abdullah <37012364+Saydemr@users.noreply.github.com> * Move URL routines to examples dir (#309) * Update blossom-ci.yml (#295) * Change pip install commands with the correct PyPI package name (#298) * reorganize * remove from init --------- Co-authored-by: Kaustubh Tangsali <71059996+ktangsali@users.noreply.github.com> Co-authored-by: Abdullah <37012364+Saydemr@users.noreply.github.com> * CorrDiff: Adding NVTX markers and removing redundant regression model inference (#306) * Refactor saving results to improve perf and overlap with inferencing * Update blossom-ci.yml (#295) * Change pip install commands with the correct PyPI package name (#298) * Make number of writer workers parameterized * add comment for writer workers * revert readme and blossom-ci * 2nd revert of readme * Adding NVTX markers and removing redundant regression model inference * Code formatting * Adding nvtx to pyproject.toml * Updating CHANGELOG.md --------- Co-authored-by: David Pruitt <dpruitt@login-eos01.eos.clusters.nvidia.com> Co-authored-by: Kaustubh Tangsali <71059996+ktangsali@users.noreply.github.com> Co-authored-by: Abdullah <37012364+Saydemr@users.noreply.github.com> Co-authored-by: David Pruitt <dpruitt@nvidia.com> Co-authored-by: David Pruitt <daviddpruitt@gmail.com> Co-authored-by: Akshay Subramaniam <asubramaniam@login-eos01.eos.clusters.nvidia.com> * Allowing channels last group norms Signed-off-by: Akshay Subramaniam <6964110+akshaysubr@users.noreply.github.com> * Adding channels last convolutions Signed-off-by: Akshay Subramaniam <6964110+akshaysubr@users.noreply.github.com> * Adding torch.compile to fix channels last groupnorm Signed-off-by: Akshay Subramaniam <6964110+akshaysubr@users.noreply.github.com> * Adding main guard in score_samples.py Signed-off-by: Akshay Subramaniam <6964110+akshaysubr@users.noreply.github.com> * Updating changelog and some cleanup Signed-off-by: Akshay Subramaniam <6964110+akshaysubr@users.noreply.github.com> * Making torch.compile configurable Signed-off-by: Akshay Subramaniam <6964110+akshaysubr@users.noreply.github.com> * Keep original group norm path for training and use custom version only for inference Signed-off-by: Akshay Subramaniam <6964110+akshaysubr@users.noreply.github.com> --------- Signed-off-by: Akshay Subramaniam <6964110+akshaysubr@users.noreply.github.com> Co-authored-by: Tao Ge <115046371+tge25@users.noreply.github.com> Co-authored-by: Mohammad Amin Nabian <m.a.nabiyan@gmail.com> Co-authored-by: Mohammad Amin Nabian <mnabian@nvidia.com> Co-authored-by: Kaustubh Tangsali <71059996+ktangsali@users.noreply.github.com> Co-authored-by: Alexey Kamenev <alex.kamenev@gmail.com> Co-authored-by: David Pruitt <dpruitt@nvidia.com> Co-authored-by: David Pruitt <dpruitt@login-eos01.eos.clusters.nvidia.com> Co-authored-by: Abdullah <37012364+Saydemr@users.noreply.github.com> Co-authored-by: David Pruitt <daviddpruitt@gmail.com> Co-authored-by: Akshay Subramaniam <asubramaniam@login-eos01.eos.clusters.nvidia.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Modulus Pull Request
Description
Make IO Asynchronous soinferencing doesn't stall while writing results
Checklist
Dependencies