Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -119,20 +119,34 @@ def analyze_text(text, model_name):
|
|
| 119 |
|
| 120 |
results = ner(text)
|
| 121 |
|
|
|
|
| 122 |
results = merge_subwords(results)
|
| 123 |
|
| 124 |
-
|
|
|
|
|
|
|
| 125 |
|
| 126 |
table_rows = []
|
| 127 |
|
| 128 |
for ent in results:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 129 |
label = ent["entity_group"]
|
| 130 |
|
| 131 |
-
|
| 132 |
-
|
| 133 |
-
|
| 134 |
-
|
| 135 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 136 |
|
| 137 |
table_rows.append([
|
| 138 |
ent["word"],
|
|
@@ -140,23 +154,113 @@ def analyze_text(text, model_name):
|
|
| 140 |
round(ent["score"], 3)
|
| 141 |
])
|
| 142 |
|
| 143 |
-
|
| 144 |
-
|
| 145 |
-
|
| 146 |
-
|
|
|
|
| 147 |
|
| 148 |
-
return
|
| 149 |
|
| 150 |
# ---------------------------------------------------
|
| 151 |
# Entity colors
|
| 152 |
# ---------------------------------------------------
|
| 153 |
|
| 154 |
COLOR_MAP = {
|
| 155 |
-
|
| 156 |
-
|
| 157 |
-
|
| 158 |
-
"
|
| 159 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 160 |
}
|
| 161 |
|
| 162 |
# ---------------------------------------------------
|
|
|
|
| 119 |
|
| 120 |
results = ner(text)
|
| 121 |
|
| 122 |
+
# merge subwords first
|
| 123 |
results = merge_subwords(results)
|
| 124 |
|
| 125 |
+
highlighted_text = []
|
| 126 |
+
|
| 127 |
+
last_idx = 0
|
| 128 |
|
| 129 |
table_rows = []
|
| 130 |
|
| 131 |
for ent in results:
|
| 132 |
+
|
| 133 |
+
start = ent["start"]
|
| 134 |
+
end = ent["end"]
|
| 135 |
+
|
| 136 |
label = ent["entity_group"]
|
| 137 |
|
| 138 |
+
# Add normal text before entity
|
| 139 |
+
if start > last_idx:
|
| 140 |
+
highlighted_text.append(
|
| 141 |
+
(text[last_idx:start], None)
|
| 142 |
+
)
|
| 143 |
+
|
| 144 |
+
# Add highlighted entity
|
| 145 |
+
highlighted_text.append(
|
| 146 |
+
(text[start:end], label)
|
| 147 |
+
)
|
| 148 |
+
|
| 149 |
+
last_idx = end
|
| 150 |
|
| 151 |
table_rows.append([
|
| 152 |
ent["word"],
|
|
|
|
| 154 |
round(ent["score"], 3)
|
| 155 |
])
|
| 156 |
|
| 157 |
+
# Add remaining text
|
| 158 |
+
if last_idx < len(text):
|
| 159 |
+
highlighted_text.append(
|
| 160 |
+
(text[last_idx:], None)
|
| 161 |
+
)
|
| 162 |
|
| 163 |
+
return highlighted_text, table_rows
|
| 164 |
|
| 165 |
# ---------------------------------------------------
|
| 166 |
# Entity colors
|
| 167 |
# ---------------------------------------------------
|
| 168 |
|
| 169 |
COLOR_MAP = {
|
| 170 |
+
# -----------------------------------
|
| 171 |
+
# Academic / theoretical
|
| 172 |
+
# -----------------------------------
|
| 173 |
+
"AcademicDiscipline": "#264653", # deep teal
|
| 174 |
+
"AmbiguouslyDefinedConcept": "#6D597A", # muted purple
|
| 175 |
+
"UnclassifiedLinguisticConcept": "#9A8C98", # soft gray-purple
|
| 176 |
+
|
| 177 |
+
# -----------------------------------
|
| 178 |
+
# Language / general linguistic
|
| 179 |
+
# -----------------------------------
|
| 180 |
+
"LanguageRelatedTerm": "#E9C46A", # warm sand yellow
|
| 181 |
+
"OtherLinguisticTerm": "#A8DADC", # pale cyan
|
| 182 |
+
"LanguageResourceInformation": "#457B9D", # medium blue
|
| 183 |
+
|
| 184 |
+
# -----------------------------------
|
| 185 |
+
# Phonology / graphemics
|
| 186 |
+
# -----------------------------------
|
| 187 |
+
"PhonologicalPhenomenon": "#E76F51", # coral red
|
| 188 |
+
"GraphemicPhenomenon": "#F4A261", # orange
|
| 189 |
+
|
| 190 |
+
# -----------------------------------
|
| 191 |
+
# Morphology / syntax
|
| 192 |
+
# -----------------------------------
|
| 193 |
+
"MorphologicalPhenomenon": "#2A9D8F", # turquoise green
|
| 194 |
+
"MorphosyntacticPhenomenon": "#52B788", # medium green
|
| 195 |
+
"SyntacticPhenomenon": "#40916C", # darker green
|
| 196 |
+
|
| 197 |
+
# -----------------------------------
|
| 198 |
+
# Lexicon / semantics / discourse
|
| 199 |
+
# -----------------------------------
|
| 200 |
+
"LexicalPhenomenon": "#577590", # slate blue
|
| 201 |
+
"SemanticPhenomenon": "#4361EE", # vivid blue
|
| 202 |
+
"DiscoursePhenomenon": "#B5179E", # magenta-purple
|
| 203 |
+
|
| 204 |
+
# -----------------------------------
|
| 205 |
+
# Special / misc
|
| 206 |
+
# -----------------------------------
|
| 207 |
+
"NEW_TAG": "#FF006E", # neon pink
|
| 208 |
+
"TOPNODE_DUMMY": "#BDBDBD", # neutral gray
|
| 209 |
+
|
| 210 |
+
# -----------------------------------
|
| 211 |
+
# Optional BIO aliases
|
| 212 |
+
# (safe fallback if model outputs raw BIO labels)
|
| 213 |
+
# -----------------------------------
|
| 214 |
+
"B-AcademicDiscipline": "#264653",
|
| 215 |
+
"I-AcademicDiscipline": "#264653",
|
| 216 |
+
|
| 217 |
+
"B-AmbiguouslyDefinedConcept": "#6D597A",
|
| 218 |
+
"I-AmbiguouslyDefinedConcept": "#6D597A",
|
| 219 |
+
|
| 220 |
+
"B-DiscoursePhenomenon": "#B5179E",
|
| 221 |
+
"I-DiscoursePhenomenon": "#B5179E",
|
| 222 |
+
|
| 223 |
+
"B-GraphemicPhenomenon": "#F4A261",
|
| 224 |
+
"I-GraphemicPhenomenon": "#F4A261",
|
| 225 |
+
|
| 226 |
+
"B-LanguageRelatedTerm": "#E9C46A",
|
| 227 |
+
"I-LanguageRelatedTerm": "#E9C46A",
|
| 228 |
+
|
| 229 |
+
"B-LanguageResourceInformation": "#457B9D",
|
| 230 |
+
"I-LanguageResourceInformation": "#457B9D",
|
| 231 |
+
|
| 232 |
+
"B-LexicalPhenomenon": "#577590",
|
| 233 |
+
"I-LexicalPhenomenon": "#577590",
|
| 234 |
+
|
| 235 |
+
"B-MorphologicalPhenomenon": "#2A9D8F",
|
| 236 |
+
"I-MorphologicalPhenomenon": "#2A9D8F",
|
| 237 |
+
|
| 238 |
+
"B-MorphosyntacticPhenomenon": "#52B788",
|
| 239 |
+
"I-MorphosyntacticPhenomenon": "#52B788",
|
| 240 |
+
|
| 241 |
+
"B-OtherLinguisticTerm": "#A8DADC",
|
| 242 |
+
"I-OtherLinguisticTerm": "#A8DADC",
|
| 243 |
+
|
| 244 |
+
"B-PhonologicalPhenomenon": "#E76F51",
|
| 245 |
+
"I-PhonologicalPhenomenon": "#E76F51",
|
| 246 |
+
|
| 247 |
+
"B-SemanticPhenomenon": "#4361EE",
|
| 248 |
+
"I-SemanticPhenomenon": "#4361EE",
|
| 249 |
+
|
| 250 |
+
"B-SyntacticPhenomenon": "#40916C",
|
| 251 |
+
"I-SyntacticPhenomenon": "#40916C",
|
| 252 |
+
|
| 253 |
+
"B-TOPNODE_DUMMY": "#BDBDBD",
|
| 254 |
+
"I-TOPNODE_DUMMY": "#BDBDBD",
|
| 255 |
+
|
| 256 |
+
"B-UnclassifiedLinguisticConcept": "#9A8C98",
|
| 257 |
+
"I-UnclassifiedLinguisticConcept": "#9A8C98",
|
| 258 |
+
|
| 259 |
+
"B-NEW_TAG": "#FF006E",
|
| 260 |
+
"I-NEW_TAG": "#FF006E",
|
| 261 |
+
|
| 262 |
+
# Outside tag
|
| 263 |
+
"O": "#FFFFFF"
|
| 264 |
}
|
| 265 |
|
| 266 |
# ---------------------------------------------------
|