This repository has been archived by the owner on Nov 16, 2023. It is now read-only.
correctness data dumps + build VM local test run script #309
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
* created correctnes branch, trimmed experiments to Dutch F3 only * trivial change to re-trigger build * dummy PR to re-trigger malfunctioning builds
* created correctnes branch, trimmed experiments to Dutch F3 only * trivial change to re-trigger build * dummy PR to re-trigger malfunctioning builds * reducing scope of the correctness branch further * added branch triggers
* upgraded to Ignite 0.3.0 and fixed upgrade compatibility * added seeds and modified notebook for ignite 0.3.0 * updated code and tests to work with ignite 0.3.0 * made code consistent with Ignite 0.3.0 as much as possible * fixed iterator epoch_length bug by subsetting validation set * applied same fix to the notebook * bugfix in distributed train.py * increased distributed tests to 2 batched - hoping for one batch per GPU * resolved rebase conflict * added seeds and modified notebook for ignite 0.3.0 * updated code and tests to work with ignite 0.3.0 * made code consistent with Ignite 0.3.0 as much as possible * fixed iterator epoch_length bug by subsetting validation set * applied same fix to the notebook * bugfix in distributed train.py * increased distributed tests to 2 batched - hoping for one batch per GPU
Co-authored-by: maxkazmsft <maxkaz@microsoft.com>
* created correctnes branch, trimmed experiments to Dutch F3 only * trivial change to re-trigger build * dummy PR to re-trigger malfunctioning builds * resolved merge conflict * flagged all non-contrib TODO with github issues * resolved rebase conflict * resolved merge conflict * cleaned up archaic voxel code
…nsborboard, bug fixes (#264) I think moving forward, we'll use smaller PRs. But here are the changes in this one: Fixes issue #236 that involves rewriting a big portion of train.py such that: All the tensorboard event handlers are organized in tensorboard_handlers.py and only called in train.py to log training and validation results in Tensorboard The code logs the same results for training and validation. Also, it adds the class IoU score as well. All single-use functions (e.g. _select_max, _tensor_to_numpy, _select_pred_and_mask) are lambda functions now The code is organized into more meaningful "chunks".. e.g. all the optimizer-related code should be together if possible, same thing for logging, configuration, loaders, tensorboard, ..etc. In addition: Fixed a visualization bug where the seismic images where not normalized correctly. This solves Issue #217. Fixed a visualization bug where the predictions where not masked where the input image was padded. This improves the ability to visually inspect and evaluate the results. This solves Issue #230. Fixes a potential issue where Tensorboard can crash when a large training batchsize is used. Now the number of images visualized in Tensorboard from every batch has an upper limit. Completely removed OpenCV as a dependency from the DeepSeismic Repo. It was only used in a small part of the code where it wasn't really necessary, and OpenCV is a huge library. Fixes Issue #218 where the epoch number for the images in Tensorboard was always logged as 1 (therefore, not allowing use to see the epoch number of the different results in Tensorboard. Removes the HorovodLRScheduler class since its no longer used Removes toolz.take from Debug mode, and uses PyTorch's native Subset() dataset class Changes default patch size for the HRNet model to 256 In addition to several other minor changes Co-authored-by: Yazeed Alaudah <yalaudah@users.noreply.github.com> Co-authored-by: Ubuntu <yazeed@yaalauda-dsvm-nd24.jsxrnelwp15e1jpgk5vvfmbzyb.bx.internal.cloudapp.net> Co-authored-by: Max Kaznady <maxkaz@microsoft.com>
* resolved rebase conflict * resolved merge conflict * resolved rebase conflict * resolved merge conflict * reverted multi-GPU builds to run on single GPU
* resolved rebase conflict * resolved merge conflict * resolved rebase conflict * resolved merge conflict * wrote the bulk of checkerboard example * finished checkerboard generator * resolved merge conflict * resolved rebase conflict * got binary dataset to run * finished first implementation mockup - commit before rebase * made sure rebase went well manually * added new files * resolved PR comments and made tests work * fixed build error * fixed build VM errors * more fixes to get the test to pass * fixed n_classes issue in data.py * fixed notebook as well * cleared notebook run cell * trivial commit to restart builds * addressed PR comments * moved notebook tests to main build pipeline * fixed checkerboard label precision * relaxed performance tests for now * resolved merge conflict * resolved merge conflict * fixed build error * resolved merge conflicts * fixed another merge mistake
* resolved rebase conflict * resolved merge conflict * resolved rebase conflict * resolved merge conflict * wrote the bulk of checkerboard example * finished checkerboard generator * resolved merge conflict * resolved rebase conflict * got binary dataset to run * finished first implementation mockup - commit before rebase * made sure rebase went well manually * added new files * resolved PR comments and made tests work * fixed build error * fixed build VM errors * more fixes to get the test to pass * fixed n_classes issue in data.py * fixed notebook as well * cleared notebook run cell * trivial commit to restart builds * addressed PR comments * moved notebook tests to main build pipeline * fixed checkerboard label precision * relaxed performance tests for now * resolved merge conflict * resolved merge conflict * fixed build error * resolved merge conflicts * fixed another merge mistake * resolved rebase conflict * resolved rebase 2 * resolved merge conflict * resolved merge conflict * adding new logging * added better logging - cleaner - debugged metrics on checkerboard dataset * resolved rebase conflict * resolved merge conflict * resolved merge conflict * resolved merge conflict * resolved rebase 2 * resolved merge conflict * updated notebook with the changes * addressed PR comments * addressed another PR comment
* correctness code good for PR review * addressed PR comments
|
Check out this pull request on Review Jupyter notebook visual diffs & provide feedback on notebooks. Powered by ReviewNB |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
closes #297 and #305