Skip to content

Return analysis for hyperparameter_search with Ray backend #22040

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
sgugger merged 5 commits into from
Mar 9, 2023
Merged

Return analysis for hyperparameter_search with Ray backend #22040

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
cd51790
return analysis for hyperparameter_search with ray backend
anruijian Mar 8, 2023
41c8b13
Revert "return analysis for hyperparameter_search with ray backend"
anruijian Mar 9, 2023
75fb1e8
add run_summary attribute to BestRun and return analysis for ray backend
anruijian Mar 9, 2023
569dbdf
fix typo
anruijian Mar 9, 2023
89bd833
add doc for run_summary for ray backend
anruijian Mar 9, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/transformers/integrations.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -346,7 +346,7 @@ def dynamic_modules_import_trainable(*args, **kwargs):
**kwargs,
)
best_trial = analysis.get_best_trial(metric="objective", mode=direction[:3], scope=trainer.args.ray_scope)
best_run = BestRun(best_trial.trial_id, best_trial.last_result["objective"], best_trial.config)
best_run = BestRun(best_trial.trial_id, best_trial.last_result["objective"], best_trial.config, analysis)
if _tb_writer is not None:
trainer.add_callback(_tb_writer)
return best_run
Expand Down
3 changes: 2 additions & 1 deletion src/transformers/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2492,7 +2492,8 @@ def hyperparameter_search(
- the documentation of [sigopt](https://app.sigopt.com/docs/endpoints/experiments/create)

Returns:
[`trainer_utils.BestRun`]: All the information about the best run.
[`trainer_utils.BestRun`]: All the information about the best run. Experiment summary can be found in
`run_summary` attribute for Ray backend.
"""
if backend is None:
backend = default_hp_search_backend()
Expand Down
5 changes: 4 additions & 1 deletion src/transformers/trainer_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -192,7 +192,7 @@ class HubStrategy(ExplicitEnum):

class BestRun(NamedTuple):
"""
The best run found by an hyperparameter search (see [`~Trainer.hyperparameter_search`]).
The best run found by a hyperparameter search (see [`~Trainer.hyperparameter_search`]).

Parameters:
run_id (`str`):
Expand All @@ -202,11 +202,14 @@ class BestRun(NamedTuple):
The objective that was obtained for this run.
hyperparameters (`Dict[str, Any]`):
The hyperparameters picked to get this run.
run_summary (`Optional[Any]`):
A summary of tuning experiments. `ray.tune.ExperimentAnalysis` object for Ray backend.
"""

run_id: str
objective: float
hyperparameters: Dict[str, Any]
run_summary: Optional[Any] = None


def default_compute_objective(metrics: Dict[str, float]) -> float:
Expand Down