microsoft/nni

nnictl experiment delete -all fail for stopped expriment

Open

#4,543 创建于 2022年2月11日

在 GitHub 查看
 (2 评论) (1 反应) (1 负责人)Python (13,504 star) (1,830 fork)batch import
good first issue

描述

Describe the issue: nnictl experiment delete -all fail for stop expriment. comment out https://github.com/microsoft/nni/blob/master/nni/tools/nnictl/nnictl_utils.py#L509 may solve this problem.

Environment: linux

  • NNI version: 2.6
  • Training service (local|remote|pai|aml|etc): local
  • Client OS: centos
  • Server OS (for remote mode only):
  • Python version: 3.7
  • PyTorch/TensorFlow version: NA
  • Is conda/virtualenv/venv used?: yes
  • Is running in Docker?: no

Configuration:

  • Experiment config (remember to remove secrets!):
  • Search space:

Log message: bash output: INFO: This action will delete experiment 74ujrqym, and it's not recoverable.
INFO: do you want to continue?[y/N]:y ERROR: There is no experiment running. ERROR: Please set correct experiment id.

How to reproduce it?:

贡献者指南

技术栈
python
领域
backend
议题类型
bug
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
2
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
under 1 hour
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
stale
清晰度议题是否清楚说明期望改动、验收标准和下一步。
mostly clear
前置要求
basic Pythonfamiliarity with nnictl
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
80
研究方向
The issue suggests that nnictl experiment delete all fails for stopped experiments. The recommended fix is to comment out line 509 in nni/tools/nnictl/nnictl utils.py which likely contains a guard that checks for a running experiment. A contributor should examine the logic around that line to ensure that deleting all experiments, including stopped ones, works correctly. There are no linked pull requests or recent comments, so this is an open task with a straightforward starting point.