Граф коммитов

6 Коммитов

Автор SHA1 Сообщение Дата
Anton Schwaighofer 5a37198b3f
Bug fix: deployed models and training code use different versions of hi-ml (#606) 2021-12-07 12:45:49 +00:00
Anton Schwaighofer bf4cb628c6
Switching batch time loading diagnostics to hi-ml (#577)
- Using the BatchTimeCallback from hi-ml
- Switches to trigger PL profiling
2021-11-03 15:48:10 +00:00
Anton Schwaighofer 7cd7e58899
Fix timeouts when downloading multiple checkpoint files (#498)
Downloading multiple checkpoints uses a codepath that has a fixed 120sec timeout. Instead, use multiple individual download operations.
2021-06-22 13:33:33 +00:00
Anton Schwaighofer 0d479ba3d8
Enable Bring-your-own-Lightning-model (#417)
- Enable brining arbitrary PyTorch-Lightning models to the InnerEye toolbox
- Upgrade mypy and simplify the way we invoke it
2021-04-19 15:28:41 +00:00
Anton Schwaighofer 95085f2d5e
Read git-related information via gitpython (#193)
Most git-related information is presently expected in commandline arguments, populated in pipelines. Change that to read via gitpython, so that also in runs from user's local machines the branch info is correct. This fixes #151
Improve documentation around queuing training runs and run recovery. Write run recovery ID to a file for later use in pipelines.
2020-09-04 14:16:03 +01:00
Shruthi42 d6a3d73ccf Add source code 2020-07-29 00:30:35 +05:30