Update pages/3_Earnings_Semantic_Search_π_.py
Browse files
pages/3_Earnings_Semantic_Search_π_.py
CHANGED
@@ -20,79 +20,85 @@ top_k = st.sidebar.slider("Number of Top Hits Generated",min_value=1,max_value=5
|
|
20 |
|
21 |
window_size = st.sidebar.slider("Number of Sentences Generated in Search Response",min_value=1,max_value=7,value=3)
|
22 |
|
23 |
-
|
24 |
-
|
25 |
-
if
|
26 |
-
|
27 |
-
## Save to a dataframe for ease of visualization
|
28 |
-
sen_df = st.session_state['sen_df']
|
29 |
-
|
30 |
-
passages = preprocess_plain_text(st.session_state['earnings_passages'],window_size=window_size)
|
31 |
-
|
32 |
-
with st.spinner(
|
33 |
-
text=f"Loading {sbert_model_name} encoder..."
|
34 |
-
):
|
35 |
-
sbert = load_sbert(sbert_model_name)
|
36 |
-
|
37 |
-
|
38 |
-
##### Sematic Search #####
|
39 |
-
# Encode the query using the bi-encoder and find potentially relevant passages
|
40 |
-
corpus_embeddings = sbert.encode(passages, convert_to_tensor=True, show_progress_bar=True)
|
41 |
-
question_embedding = sbert.encode(search_input, convert_to_tensor=True)
|
42 |
-
question_embedding = question_embedding.cpu()
|
43 |
-
hits = util.semantic_search(question_embedding, corpus_embeddings, top_k=top_k,score_function=util.dot_score)
|
44 |
-
hits = hits[0] # Get the hits for the first query
|
45 |
-
|
46 |
-
##### Re-Ranking #####
|
47 |
-
# Now, score all retrieved passages with the cross_encoder
|
48 |
-
cross_inp = [[search_input, passages[hit['corpus_id']]] for hit in hits]
|
49 |
-
cross_scores = cross_encoder.predict(cross_inp)
|
50 |
-
|
51 |
-
# Sort results by the cross-encoder scores
|
52 |
-
for idx in range(len(cross_scores)):
|
53 |
-
hits[idx]['cross-score'] = cross_scores[idx]
|
54 |
-
|
55 |
-
# Output of top-3 hits from re-ranker
|
56 |
-
hits = sorted(hits, key=lambda x: x['cross-score'], reverse=True)
|
57 |
|
58 |
-
|
59 |
-
df = pd.DataFrame([(hit[score],passages[hit['corpus_id']]) for hit in hits[0:int(top_k)]],columns=['Score','Text'])
|
60 |
-
df['Score'] = round(df['Score'],2)
|
61 |
-
df['Sentiment'] = df.Text.apply(gen_sentiment)
|
62 |
|
63 |
-
|
64 |
-
|
|
|
|
|
65 |
|
66 |
-
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
if label == 'Positive':
|
71 |
-
tag_list.append((text,label,'#8fce00'))
|
72 |
-
elif label == 'Negative':
|
73 |
-
tag_list.append((text,label,'#f44336'))
|
74 |
-
else:
|
75 |
-
tag_list.append((text,label,'#000000'))
|
76 |
|
77 |
-
return tag_list
|
78 |
-
|
79 |
-
text_annotations = gen_annotated_text(df)
|
80 |
-
|
81 |
-
first, second = text_annotations[0], text_annotations[1]
|
82 |
-
|
83 |
-
|
84 |
-
with st.expander(label='Best Search Query Result', expanded=True):
|
85 |
-
annotated_text(first)
|
86 |
|
87 |
-
|
88 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
89 |
|
90 |
else:
|
|
|
|
|
91 |
|
92 |
-
|
93 |
-
|
94 |
-
|
95 |
-
|
96 |
-
st.write('Please ensure you have entered the YouTube URL or uploaded the Earnings Call file')
|
97 |
|
98 |
|
|
|
20 |
|
21 |
window_size = st.sidebar.slider("Number of Sentences Generated in Search Response",min_value=1,max_value=7,value=3)
|
22 |
|
23 |
+
try:
|
24 |
+
|
25 |
+
if search_input:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
26 |
|
27 |
+
if "sen_df" in st.session_state and "earnings_passages" in st.session_state:
|
|
|
|
|
|
|
28 |
|
29 |
+
## Save to a dataframe for ease of visualization
|
30 |
+
sen_df = st.session_state['sen_df']
|
31 |
+
|
32 |
+
passages = preprocess_plain_text(st.session_state['earnings_passages'],window_size=window_size)
|
33 |
|
34 |
+
with st.spinner(
|
35 |
+
text=f"Loading {sbert_model_name} encoder..."
|
36 |
+
):
|
37 |
+
sbert = load_sbert(sbert_model_name)
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
39 |
|
40 |
+
##### Sematic Search #####
|
41 |
+
# Encode the query using the bi-encoder and find potentially relevant passages
|
42 |
+
corpus_embeddings = sbert.encode(passages, convert_to_tensor=True, show_progress_bar=True)
|
43 |
+
question_embedding = sbert.encode(search_input, convert_to_tensor=True)
|
44 |
+
question_embedding = question_embedding.cpu()
|
45 |
+
hits = util.semantic_search(question_embedding, corpus_embeddings, top_k=top_k,score_function=util.dot_score)
|
46 |
+
hits = hits[0] # Get the hits for the first query
|
47 |
+
|
48 |
+
##### Re-Ranking #####
|
49 |
+
# Now, score all retrieved passages with the cross_encoder
|
50 |
+
cross_inp = [[search_input, passages[hit['corpus_id']]] for hit in hits]
|
51 |
+
cross_scores = cross_encoder.predict(cross_inp)
|
52 |
+
|
53 |
+
# Sort results by the cross-encoder scores
|
54 |
+
for idx in range(len(cross_scores)):
|
55 |
+
hits[idx]['cross-score'] = cross_scores[idx]
|
56 |
+
|
57 |
+
# Output of top-3 hits from re-ranker
|
58 |
+
hits = sorted(hits, key=lambda x: x['cross-score'], reverse=True)
|
59 |
+
|
60 |
+
score='cross-score'
|
61 |
+
df = pd.DataFrame([(hit[score],passages[hit['corpus_id']]) for hit in hits[0:int(top_k)]],columns=['Score','Text'])
|
62 |
+
df['Score'] = round(df['Score'],2)
|
63 |
+
df['Sentiment'] = df.Text.apply(gen_sentiment)
|
64 |
+
|
65 |
+
def gen_annotated_text(df):
|
66 |
+
'''Generate annotated text'''
|
67 |
+
|
68 |
+
tag_list=[]
|
69 |
+
for row in df.itertuples():
|
70 |
+
label = row[3]
|
71 |
+
text = row[2]
|
72 |
+
if label == 'Positive':
|
73 |
+
tag_list.append((text,label,'#8fce00'))
|
74 |
+
elif label == 'Negative':
|
75 |
+
tag_list.append((text,label,'#f44336'))
|
76 |
+
else:
|
77 |
+
tag_list.append((text,label,'#000000'))
|
78 |
+
|
79 |
+
return tag_list
|
80 |
+
|
81 |
+
text_annotations = gen_annotated_text(df)
|
82 |
+
|
83 |
+
first, second = text_annotations[0], text_annotations[1]
|
84 |
+
|
85 |
+
|
86 |
+
with st.expander(label='Best Search Query Result', expanded=True):
|
87 |
+
annotated_text(first)
|
88 |
+
|
89 |
+
with st.expander(label='Alternative Search Query Result'):
|
90 |
+
annotated_text(second)
|
91 |
+
|
92 |
+
else:
|
93 |
+
|
94 |
+
st.write('Please ensure you have entered the YouTube URL or uploaded the Earnings Call file')
|
95 |
|
96 |
else:
|
97 |
+
|
98 |
+
st.write('Please ensure you have entered the YouTube URL or uploaded the Earnings Call file')
|
99 |
|
100 |
+
except RuntimeError:
|
101 |
+
|
102 |
+
st.write('Please ensure you have entered the YouTube URL or uploaded the Earnings Call file'
|
|
|
|
|
103 |
|
104 |
|