ashleve / lightning-hydra-template

PyTorch Lightning + Hydra. A very user-friendly template for ML experimentation. ⚡🔥⚡

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

please support dynamically selecting config_name

YChienHung opened this issue · comments

I want to use the function of dynamically selecting config_name.

Can this function be added in future versions?

Seems that hydra does not support dynamically selecting config_name. facebookresearch/hydra#2868 (comment)

What do you mean by "dynamically"?

  1. You can specify the config name in the CLI (as I comment in the quoted issue).

  2. You can also use the following: https://hydra.cc/docs/advanced/unit_testing/
    Which let you specify the config_name inside the python function (that's how I test functions that were wrapped with hydra's api).

What do you mean by "dynamically"?

  1. You can specify the config name in the CLI (as I comment in the quoted issue).
  2. You can also use the following: https://hydra.cc/docs/advanced/unit_testing/
    Which let you specify the config_name inside the python function (that's how I test functions that were wrapped with hydra's api).

Could you give me a link for solution 1 ? thanks !

What do you mean by "dynamically"?

  1. You can specify the config name in the CLI (as I comment in the quoted issue).
  2. You can also use the following: https://hydra.cc/docs/advanced/unit_testing/
    Which let you specify the config_name inside the python function (that's how I test functions that were wrapped with hydra's api).

Could you give me a link for solution 1 ? thanks !

Sure, here

What do you mean by "dynamically"?

  1. You can specify the config name in the CLI (as I comment in the quoted issue).
  2. You can also use the following: https://hydra.cc/docs/advanced/unit_testing/
    Which let you specify the config_name inside the python function (that's how I test functions that were wrapped with hydra's api).

Could you give me a link for solution 1 ? thanks !

Sure, here

thank you so much !