MetaCell / MetaCell/NetPyNE-UI

Observed issues in Experiments feature

未关闭
#354 1 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

还没有人认领这个 Issue。

主要语言
JavaScript
星标
23
派生
10
PR 合并指标
30 天内没有已合并 PR

描述

**Frontend**

Bugs

* [x] Edge case showing error "utils" is not defined for getExperiments when kernel crashed or sth bad happened (#381) .

* [x] How to handle Experiment with no parameters (no trials but one trial) in Launch Dialog? "All Trials" will trigger batch simulation but we don't have any exploration parameters. Should only allow to select "Base Model"?

* [x] Simulate button won't show LaunchDialog, only Model > Simulate

Improvements

* [x] JSONViewer performance issues with larger JSON files (#385)

* [x] Improve form validation - for existing params when editing an experiment, validation is not active because "field" wasn't fetched yet (#304)

* [ ] Available experiment parameters list should be updated when model changes (#386)

* [x] A user can create an arbitrary large number of trials with range type -> have to restrict that (#387)

**Backend**

Bugs

* [x] mpi - not enough slots available to statisfy 9 slots, although my machine has 16 cores (#372)

* [x] don't allow more than one experiment in design, have to delete existing experiment in design when we create a new one, currently just adding more experiments in design

* [x] Whitespaces in filenames cause issues for batch simulations (#384).

* [x] When we run a batch simulation and wait for the subprocess to complete, it never finishes. Some speciality with mpiexec? -> For now, all experiments will run non-blocking

Improvements

* [x] EditExperiment is inefficient because it loads all trials, have to load trials only in view experiment (#388)
* [x] Successful simulation exists only when writing of data file is completed (or file exists) - Currently, using workaround to wait for data files since exit of a simulation doesn't mean that the file exists, write operation can be delayed (#383)

**Netpyne**

* [ ] `netpyne.metadata` is not up-to-date, type doesn't seem right in some cases (float vs list) and some fields are missing (#377)

贡献指南

这个仓库没有索引到贡献指南

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

调研方向

检查两个未勾选的项目:在模型发生变化时更新可用的实验参数列表,并修正 `netpyne.metadata` 的类型和缺失字段。从相关的实验 Launch Dialog/模型变更流程以及元数据定义或使用方入手;当这两种行为都得到处理,并且引用的 issue (#386 和 #377) 可以关闭时,即表示完成。

由索引模型根据 Issue 内容生成。

评估

技术栈
javascript
领域
backend, frontend
Issue 类型
缺陷
难度
5/5
预计耗时
一周以上
活跃度
停滞
描述清晰度
需要澄清
新手友好度
15/100

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。