Skip to content

[Fix] Fix eval #931

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 7 commits into from
Jun 21, 2024
Merged

[Fix] Fix eval #931

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
1d250f0
update develop mkdocs
HydrogenSulfate Jun 19, 2024
cb28eab
Merge branch 'develop' of https://github.com/HydrogenSulfate/PaddleSc…
HydrogenSulfate Jun 19, 2024
8e38247
allow alias for mike
HydrogenSulfate Jun 19, 2024
167b789
Merge branch 'develop' of https://github.com/PaddlePaddle/PaddleScien…
HydrogenSulfate Jun 19, 2024
347f516
fix and refine eval.py
HydrogenSulfate Jun 20, 2024
246120c
Merge branch 'develop' into fix_eval
HydrogenSulfate Jun 20, 2024
ce6a983
fix
HydrogenSulfate Jun 20, 2024
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
16 changes: 11 additions & 5 deletions ppsci/solver/eval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -167,6 +167,10 @@ def _eval_by_dataset(
for metric_name, metric_func in _validator.metric.items():
# NOTE: compute metric with entire output and label
metric_dict = metric_func(all_output, all_label)
assert metric_name not in metric_dict_group, (
f"Metric name({metric_name}) already exists, please ensure all metric "
"names are unique over all validators."
)
metric_dict_group[metric_name] = {
k: float(v) for k, v in metric_dict.items()
}
Expand Down Expand Up @@ -215,7 +219,6 @@ def _eval_by_batch(
num_samples = _get_dataset_length(_validator.data_loader)

loss_dict = misc.Prettydefaultdict(float)
metric_dict_group: Dict[str, Dict[str, float]] = misc.PrettyOrderedDict()
reader_tic = time.perf_counter()
batch_tic = time.perf_counter()
for iter_id, batch in enumerate(_validator.data_loader, start=1):
Expand Down Expand Up @@ -251,9 +254,12 @@ def _eval_by_batch(

# collect batch metric
for metric_name, metric_func in _validator.metric.items():
assert metric_name not in metric_dict_group, (
f"Metric name({metric_name}) already exists, please ensure all metric "
"names are unique over all validators."
)
metric_dict_group[metric_name] = misc.Prettydefaultdict(list)
metric_dict = metric_func(output_dict, label_dict)
if metric_name not in metric_dict_group:
metric_dict_group[metric_name] = misc.Prettydefaultdict(list)
for var_name, metric_value in metric_dict.items():
metric_dict_group[metric_name][var_name].append(
metric_value
Expand Down Expand Up @@ -284,9 +290,9 @@ def _eval_by_batch(
# concatenate all metric and discard metric of padded sample(s)
for metric_name, metric_dict in metric_dict_group.items():
for var_name, metric_value in metric_dict.items():
# NOTE: concat all metric(scalars) into metric vector
# NOTE: concat single metric(scalar) list into metric vector
metric_value = paddle.concat(metric_value)[:num_samples]
# NOTE: compute metric via averaging metric vector,
# NOTE: compute metric via averaging metric over all samples,
# this might be not general for certain evaluation case
metric_value = float(metric_value.mean())
metric_dict_group[metric_name][var_name] = metric_value
Expand Down