Update app.py
Browse files
app.py
CHANGED
@@ -87,20 +87,22 @@ def process_text(inputs: str, confidence_threshold: float):
|
|
87 |
for word_start in sorted(word_info.keys()):
|
88 |
word_data = word_info[word_start]
|
89 |
for subtoken_start, subtoken_end, subtoken_text in word_data['subtokens']:
|
|
|
90 |
if last_end < subtoken_start:
|
91 |
reconstructed_text += chunk[last_end:subtoken_start]
|
92 |
if word_data['prediction'] == 1:
|
93 |
-
reconstructed_text += f"<span style='background-color: rgba(0, 255, 0); display: inline;'>{
|
94 |
else:
|
95 |
-
reconstructed_text +=
|
96 |
last_end = subtoken_end
|
97 |
|
98 |
-
df_data['Word'].append(
|
99 |
df_data['Prediction'].append(word_data['prediction'])
|
100 |
df_data['Confidence'].append(word_info[word_start]['confidence'])
|
101 |
df_data['Start'].append(subtoken_start + original_position_offset)
|
102 |
df_data['End'].append(subtoken_end + original_position_offset)
|
103 |
|
|
|
104 |
original_position_offset += len(chunk) + 1
|
105 |
|
106 |
reconstructed_text += chunk[last_end:].replace('$', '\\$')
|
|
|
87 |
for word_start in sorted(word_info.keys()):
|
88 |
word_data = word_info[word_start]
|
89 |
for subtoken_start, subtoken_end, subtoken_text in word_data['subtokens']:
|
90 |
+
escaped_subtoken_text = subtoken_text.replace('$', '\\$') # Perform replacement outside f-string
|
91 |
if last_end < subtoken_start:
|
92 |
reconstructed_text += chunk[last_end:subtoken_start]
|
93 |
if word_data['prediction'] == 1:
|
94 |
+
reconstructed_text += f"<span style='background-color: rgba(0, 255, 0); display: inline;'>{escaped_subtoken_text}</span>"
|
95 |
else:
|
96 |
+
reconstructed_text += escaped_subtoken_text
|
97 |
last_end = subtoken_end
|
98 |
|
99 |
+
df_data['Word'].append(escaped_subtoken_text)
|
100 |
df_data['Prediction'].append(word_data['prediction'])
|
101 |
df_data['Confidence'].append(word_info[word_start]['confidence'])
|
102 |
df_data['Start'].append(subtoken_start + original_position_offset)
|
103 |
df_data['End'].append(subtoken_end + original_position_offset)
|
104 |
|
105 |
+
|
106 |
original_position_offset += len(chunk) + 1
|
107 |
|
108 |
reconstructed_text += chunk[last_end:].replace('$', '\\$')
|