SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Update docs to explain disabling callbacks using report_to (#26155)

Browse Source 
* feat: update callback doc to explain disabling callbacks using report_to

* docs: update report_to docstring
This commit is contained in:
Ben Gubler
2023-10-11 05:50:23 -06:00
committed by GitHub
parent dcc49d8a7e
commit 9f40639292
2 changed files with 7 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
docs/source/en/main_classes/callback.md
View File

@@ -25,7 +25,7 @@ Callbacks are "read only" pieces of code, apart from the [`TrainerControl`] obje
cannot change anything in the training loop. For customizations that require changes in the training loop, you should cannot change anything in the training loop. For customizations that require changes in the training loop, you should
subclass [`Trainer`] and override the methods you need (see [trainer](trainer) for examples). subclass [`Trainer`] and override the methods you need (see [trainer](trainer) for examples).
By default a [`Trainer`] will use the following callbacks: By default, `TrainingArguments.report_to` is set to `"all"`, so a [`Trainer`] will use the following callbacks.
- [`DefaultFlowCallback`] which handles the default behavior for logging, saving and evaluation. - [`DefaultFlowCallback`] which handles the default behavior for logging, saving and evaluation.
- [`PrinterCallback`] or [`ProgressCallback`] to display progress and print the - [`PrinterCallback`] or [`ProgressCallback`] to display progress and print the
@@ -45,6 +45,8 @@ By default a [`Trainer`] will use the following callbacks:
- [`~integrations.DagsHubCallback`] if [dagshub](https://dagshub.com/) is installed. - [`~integrations.DagsHubCallback`] if [dagshub](https://dagshub.com/) is installed.
- [`~integrations.FlyteCallback`] if [flyte](https://flyte.org/) is installed. - [`~integrations.FlyteCallback`] if [flyte](https://flyte.org/) is installed.
If a package is installed but you don't wish to use the accompanying integration, you can change `TrainingArguments.report_to` to a list of just those integrations you want to use (e.g. `["azure_ml", "wandb"]`).
The main class that implements callbacks is [`TrainerCallback`]. It gets the The main class that implements callbacks is [`TrainerCallback`]. It gets the
[`TrainingArguments`] used to instantiate the [`Trainer`], can access that [`TrainingArguments`] used to instantiate the [`Trainer`], can access that
Trainer's internal state via [`TrainerState`], and can take some actions on the training loop via Trainer's internal state via [`TrainerState`], and can take some actions on the training loop via

7
src/transformers/training_args.py
View File

@@ -2345,10 +2345,11 @@ class TrainingArguments:
Logger log level to use on the main process. Possible choices are the log levels as strings: `"debug"`, Logger log level to use on the main process. Possible choices are the log levels as strings: `"debug"`,
`"info"`, `"warning"`, `"error"` and `"critical"`, plus a `"passive"` level which doesn't set anything `"info"`, `"warning"`, `"error"` and `"critical"`, plus a `"passive"` level which doesn't set anything
and lets the application set the level. and lets the application set the level.
report_to (`str` or `List[str]`, *optional*, defaults to `"none"`): report_to (`str` or `List[str]`, *optional*, defaults to `"all"`):
The list of integrations to report the results and logs to. Supported platforms are `"azure_ml"`, The list of integrations to report the results and logs to. Supported platforms are `"azure_ml"`,
`"comet_ml"`, `"mlflow"`, `"neptune"`, `"tensorboard"`,`"clearml"` and `"wandb"`. Use `"all"` to report `"clearml"`, `"codecarbon"`, `"comet_ml"`, `"dagshub"`, `"flyte"`, `"mlflow"`, `"neptune"`,
to all integrations installed, `"none"` for no integrations. `"tensorboard"`, and `"wandb"`. Use `"all"` to report to all integrations installed, `"none"` for no
integrations.
first_step (`bool`, *optional*, defaults to `False`): first_step (`bool`, *optional*, defaults to `False`):
Whether to log and evaluate the first `global_step` or not. Whether to log and evaluate the first `global_step` or not.
nan_inf_filter (`bool`, *optional*, defaults to `True`): nan_inf_filter (`bool`, *optional*, defaults to `True`):