Spaces:
Sleeping
Sleeping
Update classification.py
Browse files- classification.py +5 -5
classification.py
CHANGED
@@ -178,17 +178,17 @@ def match_categories(df, category_df, treshold=0.45):
|
|
178 |
if isinstance(ebd_content, torch.Tensor):
|
179 |
cos_scores = util.cos_sim(ebd_content, torch.stack(list(category_df['Embeddings']), dim=0))[0]
|
180 |
high_score_indices = [i for i, score in enumerate(cos_scores) if score > treshold]
|
181 |
-
categories_list.append([category_df.loc[index, 'description'] for index in high_score_indices])
|
182 |
-
experts_list.append([category_df.loc[index, 'experts'] for index in high_score_indices])
|
183 |
-
topic_list.append([category_df.loc[index, 'topic'] for index in high_score_indices])
|
184 |
-
scores_list.append([float(cos_scores[index]) for index in high_score_indices])
|
185 |
for j in high_score_indices:
|
186 |
df.loc[index, category_df.loc[j, 'topic']] = float(cos_scores[j])
|
187 |
else:
|
188 |
categories_list.append(np.nan)
|
189 |
experts_list.append(np.nan)
|
190 |
topic_list.append(np.nan)
|
191 |
-
scores_list.append(
|
192 |
df["Description"] = categories_list
|
193 |
df["Expert"] = experts_list
|
194 |
df["Topic"] = topic_list
|
|
|
178 |
if isinstance(ebd_content, torch.Tensor):
|
179 |
cos_scores = util.cos_sim(ebd_content, torch.stack(list(category_df['Embeddings']), dim=0))[0]
|
180 |
high_score_indices = [i for i, score in enumerate(cos_scores) if score > treshold]
|
181 |
+
categories_list.append("@~@".join([category_df.loc[index, 'description'] for index in high_score_indices]))
|
182 |
+
experts_list.append("@~@".join([category_df.loc[index, 'experts'] for index in high_score_indices]))
|
183 |
+
topic_list.append("@~@".join([category_df.loc[index, 'topic'] for index in high_score_indices]))
|
184 |
+
scores_list.append("@~@".join([float(cos_scores[index]) for index in high_score_indices]))
|
185 |
for j in high_score_indices:
|
186 |
df.loc[index, category_df.loc[j, 'topic']] = float(cos_scores[j])
|
187 |
else:
|
188 |
categories_list.append(np.nan)
|
189 |
experts_list.append(np.nan)
|
190 |
topic_list.append(np.nan)
|
191 |
+
scores_list.append(np.nan)
|
192 |
df["Description"] = categories_list
|
193 |
df["Expert"] = experts_list
|
194 |
df["Topic"] = topic_list
|