This repository has been archived by the owner on Nov 16, 2023. It is now read-only.
V0.1.2 #307
Merged
V0.1.2 #307
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 Python package structure
- Created Python package structure for generative models for velocities - Implemented the [Röth-Tarantola model](https://doi.org/10.1029/93JB01563)
Implemented forward modelling for the isotropic acoustic wave equation using [Devito](https://www.devitoproject.org/)
Exposed PRNG seed in generative models for velocities
- Updated LICENSE - Added Microsoft Open Source Code of Conduct - Added Contributing section to README
Implemented CLI for velocity generators
Reimplemented CLI as subpackage using Click
Added VS Code settings
Implemented CLI for forward modelling
- Changed to use km/s instead of m/s for velocities - Fixed CLI interface
Basic infrastructure PR.
Adding the cgmanifest.json file to staging branch, so I could create a build task to run the Component Governance on staging branch.
…opment/_git/DeepSeismic into vapaunic/cgmanifest
A PR to add prelim NOTICE.txt file.
* added sharat, weehyong to authors * adding a download script for Dutch F3 dataset * Adding script instructions for dutch f3 * Update README.md prepare scripts expect root level directory for dutch f3 dataset. (it is downloaded into $dir/data by the script) * added scripts which reproduce results * Adding readme text for the notebooks and checking if config is correctly setup * build error fix * fixed notebook build error * removed scripts from older pull * fixing pip version for DS VM bug fix * fixing pip version for DS VM bug fix * notebook fixes * fixes to benchmark test script * explicitly setting backbone model location in tests * addressed PR comments * updated model test script to work on master and staging branches to reproduce results * enables ability to change notebook execution dir * fix Co-authored-by: Sharat Chikkerur <sharat.chikkerur@gmail.com>
* removed duplicate code in the notebooks * initial draft * done with download_pretrained_model * updated notebook and utils * updating model dir in config * updates to util * update to notebook * model download fixes and HRNet pre-trained model demo run * fix to non-existant model_dir directory on the build server * typo Co-authored-by: maxkazmsft <maxkaz@microsoft.com>
* added ability to use pre-trained models on Dutch F3 dataset * moved black notebook formatter instructions to README * finished Dutch F3 notebook training - pre trained model runtime is down to 1 minute; starting on test set performance * finished dutch f3 notebook * fixed Docker not running out-of-the-box with the given parameters * cleaned up other notebooks and files which are not scoped for this release * tweaks to notebook from Docker * fixed Docker instructions and port 9000 for TB * notebook build fixes * small Dockerfile fix * notebook build fixes * increased max_iterations in tests * finished tweaking the notebook to get the tests to pass * more fixes for build tests * dummy commit to re-trigger the builds * addressed PR comments * reverting back data.Subset to toolz.take
* added docker image test build * increased Docker image build timeout
…#246) * re-wrote experiment test builds to run in parallel on single 4-GPU VM * fixed yaml typo * fixed another yaml typo * added more descriptive build names * fixed another yaml typo * changed build names and added tee log splitting * added wait -n * added wait termination condition * fixed path typo * added code to manually block on PIDs * added ADO fixes to collect PIDs for wait; changed component governance build pool * added manual handling of return codes * fixed parallel distributed tests * build typo
* 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
…ng into correctness
* 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
* updated readme for v0.2 release
|
Check out this pull request on Review Jupyter notebook visual diffs & provide feedback on notebooks. Powered by ReviewNB |
sharatsc
approved these changes
May 20, 2020
yalaudah
approved these changes
May 20, 2020
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.
No description provided.