Skip to content
This repository has been archived by the owner on Mar 21, 2024. It is now read-only.

Issues: microsoft/InnerEye-DeepLearning

Clear current search query, filters, and sorts
84 Open 212 Closed
84 Open 212 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Restarting training from checkpoint on large model goes out of memory bug Something isn't working
#351 opened Jan 5, 2021 by javier-alvarez updated May 19, 2022 Run recovery and pre-emption recovery work fine
@peterhessey
1
Patch visualization: Highlight that areas that are never sampled good first issue Good for newcomers not urgent Not needed for current work reporting and diagnostics Improvements to diagnostics and reporting
#301 opened Oct 28, 2020 by ant0nsc updated May 19, 2022
Suggestion to include SEG and not only RTSTRUCT IODs for the segmentations deployment Work to enable/improve model deployment question Further information is requested usage blocker Bugs that block InnerEye users
#382 opened Jan 29, 2021 by athanell updated May 19, 2022
@peterhessey
1
Retire run recovery ID, replace with RunID good first issue Good for newcomers
#571 opened Sep 30, 2021 by ant0nsc updated May 19, 2022
@peterhessey
Re-enable test_recovery_on_2_nodes bug Something isn't working question Further information is requested workflow Improvements to workflows (recovery, speed)
#556 opened Aug 25, 2021 by Shruthi42 updated May 19, 2022
@peterhessey
Add a "download files" script that recognizes patterns usage blocker Bugs that block InnerEye users
#524 opened Jul 6, 2021 by ant0nsc updated May 19, 2022
@peterhessey
2
Benchmark the evaluation of a segmentation model on a full test set image not urgent Not needed for current work
#479 opened Jun 7, 2021 by ant0nsc updated May 19, 2022
1
Fallback run for tests failed and it makes tests fail locally bug Something isn't working
#665 opened Feb 18, 2022 by fepegar updated May 19, 2022
@peterhessey
Extend submit_for_inference to work correctly with hi-ml as submodule
#694 opened Mar 10, 2022 by ant0nsc updated Mar 10, 2022
Move SSL models to HI-ML and delete from InnerEye
#691 opened Mar 9, 2022 by ant0nsc updated Mar 9, 2022
Test-requirements.txt uses differnt version of pl-bolts than main environment
#685 opened Mar 3, 2022 by ant0nsc updated Mar 3, 2022
Error during inference if patch size is larger than image bug Something isn't working
#676 opened Mar 1, 2022 by fepegar updated Mar 1, 2022
2 tasks
@fepegar
1
Average overlapping predictions in patch-based inference
#673 opened Feb 28, 2022 by fepegar updated Feb 28, 2022
@fepegar
Add config parameter to fail training run if model comparison fails.
#668 opened Feb 24, 2022 by javier-alvarez updated Feb 25, 2022
@fepegar
Pytorch-lightning < 1.5.8 is vulnerable and we are using 1.5.5
#658 opened Feb 15, 2022 by fepegar updated Feb 15, 2022
Enable the new recovery functionality of PL 1.5
#627 opened Jan 5, 2022 by ant0nsc updated Jan 10, 2022
1
@ant0nsc
1
Dataset mounting on local VMs points to wrong folders
#624 opened Dec 15, 2021 by ant0nsc updated Dec 15, 2021
@ant0nsc
1
Deleting folders from job output can cause trouble with hosttools
#610 opened Dec 7, 2021 by ant0nsc updated Dec 7, 2021
SSL: What are the newest/best SSL implementations that are available?
#600 opened Dec 1, 2021 by ant0nsc updated Dec 1, 2021
SSL: Enable custom transformations that extend the present set of transformations
#598 opened Dec 1, 2021 by ant0nsc updated Dec 1, 2021
1
SSL: metadata is removed and not fully available in training_step methods
#597 opened Dec 1, 2021 by ant0nsc updated Dec 1, 2021
Check consistency of SSL metrics on multiple nodes
#595 opened Nov 24, 2021 by ant0nsc updated Nov 24, 2021
@ant0nsc
Enable building an ensemble model from the cross validation checkpoints of a BYOL model
#527 opened Jul 6, 2021 by dumbledad updated Nov 18, 2021
2
1
Multi-GPU jobs don't terminate when one worker fails
#590 opened Nov 17, 2021 by ant0nsc updated Nov 17, 2021
Jobs file when they get pre-empted after uploading files
#568 opened Sep 27, 2021 by ant0nsc updated Sep 27, 2021
1
@JonathanTripp
ProTip! Mix and match filters to narrow down what you’re looking for.