- 有参数的debug
这个很容易做到,通过launch.json就可以解决。
- 选择下图第四个
- 新建launch.json
- 主要是添加args
{ "version": "0.2.0", "configurations": [ { "name": "Python: Current File" "type": "python", "request": "launch", "program": "${file}", "console": "integratedTerminal", "args": [ "arg1", "xxx", "arg2", "xxx", ] } }
- 分布式训练debug
这个相比只有参数的debug复杂了一点,但是也是只有launch.json的区别:{ "version": "0.2.0", "configurations": [ { "name": "Python: Current File", "type": "python", "request": "launch", "console": "integratedTerminal", "args": [ "--nproc_per_node=2", "main.py", ], "env": {"CUDA_VISIBLE_DEVICES":"0,1"}, "program": "/home/lucas/anaconda3/envs/py39-pt112-cu113/lib/python3.9/site-packages/torch/distributed/run.py", } ] }