Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -156,15 +156,6 @@ def predict_test(model, tokenizer, embedding_model, df, question, index): # sen
|
|
156 |
mostSimContext = re.sub(r'\s+', ' ', mostSimContext)
|
157 |
|
158 |
segments = sent_tokenize(mostSimContext, engine="crfcut")
|
159 |
-
|
160 |
-
#=====add
|
161 |
-
if (len(segments)==1):
|
162 |
-
segments = ' '.join(segments)
|
163 |
-
|
164 |
-
segments = segments.split('และ')
|
165 |
-
segments = [segment.split('หรือ') for segment in segments]
|
166 |
-
segments = [sentence for segment in segments for sentence in segment]
|
167 |
-
#=====end
|
168 |
|
169 |
segments_index = set_index(get_embeddings(embedding_model,segments))
|
170 |
_distances,_indices = faiss_search(segments_index, question_vector)
|
|
|
156 |
mostSimContext = re.sub(r'\s+', ' ', mostSimContext)
|
157 |
|
158 |
segments = sent_tokenize(mostSimContext, engine="crfcut")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
159 |
|
160 |
segments_index = set_index(get_embeddings(embedding_model,segments))
|
161 |
_distances,_indices = faiss_search(segments_index, question_vector)
|