Skip to content

callback: use logging instance #2

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
merged 1 commit into from
Dec 24, 2024
Merged

callback: use logging instance #2

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
22 changes: 11 additions & 11 deletions d3rlpy/algos/qlearning/base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -508,7 +508,7 @@ def fitter(
# setup logger
if experiment_name is None:
experiment_name = self.__class__.__name__
logger = D3RLPyLogger(
self.logger = D3RLPyLogger(
algo=self,
adapter_factory=logger_adapter,
experiment_name=experiment_name,
Expand All @@ -517,7 +517,7 @@ def fitter(
)

# save hyperparameters
save_config(self, logger)
save_config(self, self.logger)

# training loop
n_epochs = n_steps // n_steps_per_epoch
Expand All @@ -533,20 +533,20 @@ def fitter(
)

for itr in range_gen:
with logger.measure_time("step"):
with self.logger.measure_time("step"):
# pick transitions
with logger.measure_time("sample_batch"):
with self.logger.measure_time("sample_batch"):
batch = dataset.sample_transition_batch(
self._config.batch_size
)

# update parameters
with logger.measure_time("algorithm_update"):
with self.logger.measure_time("algorithm_update"):
loss = self.update(batch)

# record metrics
for name, val in loss.items():
logger.add_metric(name, val)
self.logger.add_metric(name, val)
epoch_loss[name].append(val)

# update progress postfix with losses
Expand All @@ -562,7 +562,7 @@ def fitter(
logging_strategy == LoggingStrategy.STEPS
and total_step % logging_steps == 0
):
metrics = logger.commit(epoch, total_step)
metrics = self.logger.commit(epoch, total_step)

# call callback if given
if callback:
Expand All @@ -575,19 +575,19 @@ def fitter(
if evaluators:
for name, evaluator in evaluators.items():
test_score = evaluator(self, dataset)
logger.add_metric(name, test_score)
self.logger.add_metric(name, test_score)

# save metrics
if logging_strategy == LoggingStrategy.EPOCH:
metrics = logger.commit(epoch, total_step)
metrics = self.logger.commit(epoch, total_step)

# save model parameters
if epoch % save_interval == 0:
logger.save_model(total_step, self)
self.logger.save_model(total_step, self)

yield epoch, metrics

logger.close()
self.logger.close()

def fit_online(
self,
Expand Down