NousResearch/atropos
1
0
Fork 0
mirror of https://github.com/NousResearch/atropos.git synced 2026-04-19 12:57:58 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Update README.md

Browse source
This commit is contained in:
crStiv 2026-01-15 16:09:46 +02:00 • committed by GitHub
parent b624cbd246
commit 7e12fa015c
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
environments/community/pytorch_optimizer_coding/FOB/pytorch_fob/tasks/README.md
View file

@ -47,7 +47,7 @@ def __init__(self, optimizer: Optimizer, config: TaskConfig):
```
In the `__init__` method you need to create your model, and pass it to the `super().__init__` call. There the model is wrapped into a `GroupedModel` which splits the model parameters into weight_decay and non-weight_decay groups. If you want to specify your own parameter groups (e.g. for different learning rates) you need to wrap your model in a `GroupedModel` yourself, before passing it to the `super().__init__` call.
The other methods you neet to implement are [training_step](https://lightning.ai/docs/pytorch/stable/common/lightning_module.html#training-step), [validation_step](https://lightning.ai/docs/pytorch/stable/common/lightning_module.html#validation-step) and [test_step](https://lightning.ai/docs/pytorch/stable/common/lightning_module.html#test-step). Here you need to implement the training and evaluation logic.
The other methods you need to implement are [training_step](https://lightning.ai/docs/pytorch/stable/common/lightning_module.html#training-step), [validation_step](https://lightning.ai/docs/pytorch/stable/common/lightning_module.html#validation-step) and [test_step](https://lightning.ai/docs/pytorch/stable/common/lightning_module.html#test-step). Here you need to implement the training and evaluation logic.
### task.py
Here you only need to provide two simple functions: