AllenCell / AllenCell/allencell-segmenter-ml

General: what happen after a crash?

未关闭
#345 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看
主要语言
Python
星标
5
派生
3
PR 合并指标
30 天内没有已合并 PR

描述

Having this ticket here to remind us to think about what should happen when napari crashes

### Scenarios
In the cases where user's device crashes (due to other apps, power loss, etc.) or the plugin/napari crashes or being unresponsive, when the user reopens napari and our plugin, at the moment, they would lose all previous work and have to start again:
- if they were in the middle of Curation, they will have to reload their image directories and restart the curation process from the beginning
- if they were in the middle of Training and no checkpoint file had been created, they might not be able to access the directory as there is no available model (?)
- if they were in the middle of Training and at least one checkpoint file had been created, they will be able to select "run an existing model" and use that checkpoint file with Prediction, but no further action is available (no Curation/Training will be available with "run an existing model" option in MVP v1)

贡献指南

这个仓库没有索引到贡献指南

评估

这个 Issue 还没有评估数据。

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。