Skip to content
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

UT for average_precision_score #803

Merged
merged 1 commit into from
Nov 1, 2023
Merged

UT for average_precision_score #803

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
42 changes: 36 additions & 6 deletions verticapy/tests_new/machine_learning/metrics/test_classification_metrics.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -252,6 +252,7 @@ def python_metrics(y_true, y_pred, average="binary", metric_name=""):
("n", "negative_likelihood_ratio"),
("n", "positive_likelihood_ratio"),
("y", "precision_score"),
("y", "average_precision_score"),
("n", "prevalence_threshold"),
("y", "recall_score"),
("n", "specificity_score"),
Expand Down Expand Up @@ -321,18 +322,30 @@ def get_vertica_metrics():
"prc_auc_score",
"best_cutoff",
"log_loss",
"average_precision_score",
]:
if compute_method == "binary":
_vpy_res = getattr(vpy_metrics, metric_name)(
"y_true", "y_prob", vdf, average=compute_method, **func_args
)
if metric_name == "average_precision_score":
_vpy_res = getattr(vpy_metrics, metric_name)(
"y_true_num",
"y_pred_num",
vdf,
average=compute_method,
pos_label="1",
)
else:
_vpy_res = getattr(vpy_metrics, metric_name)(
"y_true", "y_prob", vdf, average=compute_method, **func_args
)
else:
_vpy_res = getattr(vpy_metrics, metric_name)(
"y_true_num",
["y_prob0", "y_prob1", "y_prob2"],
vdf,
average=compute_method,
labels=labels_num,
labels=[str(label_num) for label_num in labels_num]
if metric_name == "average_precision_score"
else labels_num,
)
# rounding as best_cutoff metrics value precisions are upto 2/3 decimals
_vpy_res = (
Expand Down Expand Up @@ -418,6 +431,18 @@ def get_python_metrics():
_y_true_num.ravel(), y_prob.ravel()
)
_skl_res = skl_metrics.auc(fpr, tpr)
elif metric_name in ["average_precision_score"]:
if compute_method == "binary":
_skl_res = getattr(skl_metrics, metric_name)(
y_true_num, y_pred_num, pos_label=1
)
else:
_skl_res = getattr(skl_metrics, metric_name)(
y_true_num,
y_prob,
average=compute_method,
pos_label=1,
)
else:
_skl_res = getattr(skl_metrics, metric_name)(
y_true, y_pred, labels=labels
Expand All @@ -435,8 +460,6 @@ def get_python_metrics():
_skl_res = skl_metrics.auc(recall, precision)
else:
y_true_num = label_binarize(y_true, classes=[0, 1, 2])
print()
print(y_true_num)
fpr, tpr, thresholds = skl_metrics.roc_curve(
y_true_num, y_prob, pos_label="b"
)
Expand All @@ -456,6 +479,13 @@ def get_python_metrics():

# rounding as best_cutoff metrics value precisions are upto 2/3 decimals
_skl_res = round(_skl_res, 2)
elif metric_name in ["average_precision_score"]:
_skl_res = getattr(skl_metrics, metric_name)(
y_true_num,
y_prob,
average=None,
pos_label=1,
)
else:
_skl_res = python_metrics(
y_true, y_pred, average=compute_method, metric_name=metric_name
Expand Down