diff options
| author | Adam Janovsky | 2023-10-20 10:29:18 +0200 |
|---|---|---|
| committer | Adam Janovsky | 2023-10-20 10:29:18 +0200 |
| commit | c7ad8444996bcbf639ef4fec0af7297374cfbfd1 (patch) | |
| tree | a833a61a5e19f87642fce44fb4dd4f8d34266573 | |
| parent | d21539ea3b6e06d3e0ba4705c740c6ef5d85e895 (diff) | |
| download | sec-certs-c7ad8444996bcbf639ef4fec0af7297374cfbfd1.tar.gz sec-certs-c7ad8444996bcbf639ef4fec0af7297374cfbfd1.tar.zst sec-certs-c7ad8444996bcbf639ef4fec0af7297374cfbfd1.zip | |
recertification -> reevaluation in code
| -rw-r--r-- | notebooks/cc/reference_annotations/prediction.ipynb | 2 | ||||
| -rw-r--r-- | src/sec_certs/model/references/annotator.py | 8 |
2 files changed, 5 insertions, 5 deletions
diff --git a/notebooks/cc/reference_annotations/prediction.ipynb b/notebooks/cc/reference_annotations/prediction.ipynb index 628f3204..fe51f611 100644 --- a/notebooks/cc/reference_annotations/prediction.ipynb +++ b/notebooks/cc/reference_annotations/prediction.ipynb @@ -101,7 +101,7 @@ "df = df.merge(all_annotations.loc[:, [\"dgst\", \"referenced_cert_id\", \"name_similarity_stripped_version\", \"name_len_diff\", \"cert_name\", \"referenced_cert_name\"]], on=[\"dgst\", \"referenced_cert_id\"])\n", "\n", "# Simplified binary labels\n", - "# label_mapping = {\"COMPONENT_USED\": \"COMPONENT_SHARED\", \"RECERTIFICATION\": \"PREVIOUS_VERSION\"}\n", + "# label_mapping = {\"COMPONENT_USED\": \"COMPONENT_SHARED\", \"REEVALUATION\": \"PREVIOUS_VERSION\"}\n", "# df.label = df.label.map(lambda x: label_mapping[x] if x in label_mapping else x)\n" ] }, diff --git a/src/sec_certs/model/references/annotator.py b/src/sec_certs/model/references/annotator.py index b7666f34..76676895 100644 --- a/src/sec_certs/model/references/annotator.py +++ b/src/sec_certs/model/references/annotator.py @@ -96,7 +96,7 @@ class ReferenceAnnotator: WIll read df.segments and populate the dataframe with predictions. """ - def matches_recertification(segments: list[str]) -> bool: + def matches_reevaluation(segments: list[str]) -> bool: regex_a = r"This is a re-?\s?certification based on (the\s){1,2}referenced product" regex_b = r"Re-?\s?Zertifizierung basierend auf (the\s){1,2}referenced product" return any( @@ -113,12 +113,12 @@ class ReferenceAnnotator: df_new.loc[ (df_new.name_similarity == 100) & (df_new.name_len_diff < 5) - & ((df_new.y_pred != "RECERTIFICATION") & (df_new.y_pred != "PREVIOUS_VERSION")), + & ((df_new.y_pred != "RE-EVALUATION") & (df_new.y_pred != "PREVIOUS_VERSION")), ["y_pred"], ] = "PREVIOUS_VERSION" - df_new["matches_recertification"] = df_new.segments.map(matches_recertification) - df_new.loc[df_new.matches_recertification, ["y_pred"]] = "RECERTIFICATION" + df_new["maches_reevaluation"] = df_new.segments.map(matches_reevaluation) + df_new.loc[df_new.maches_reevaluation, ["y_pred"]] = "RE-EVALUATION" df_new["correct"] = df_new.apply( lambda row: row["y_pred"] == row["label"] if not pd.isnull(row["label"]) else np.NaN, axis=1 |