Spaces:
Paused
Paused
Update app.py
Browse files
app.py
CHANGED
|
@@ -36,12 +36,13 @@ with col2:
|
|
| 36 |
option = st.selectbox(
|
| 37 |
'Select parsing method.',
|
| 38 |
('monkey', 'x2d', 'lxml'))
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
|
|
|
| 45 |
|
| 46 |
summ_text = None
|
| 47 |
|
|
@@ -73,6 +74,7 @@ if (uploaded_file is not None) and (not 'generation_done' in st.session_state):
|
|
| 73 |
|
| 74 |
with st.status("Generating slides..."):
|
| 75 |
summarizor = pipeline("summarization", model=summ_model_path, device = device)
|
|
|
|
| 76 |
summ_text = summarizor(document.segmentation['text'], max_length=100, min_length=10, do_sample=False)
|
| 77 |
summ_text = [text['summary_text'] for text in summ_text]
|
| 78 |
|
|
@@ -81,14 +83,15 @@ if (uploaded_file is not None) and (not 'generation_done' in st.session_state):
|
|
| 81 |
if (summ_text is not None) or ('summ_text' in st.session_state):
|
| 82 |
|
| 83 |
# Function to render HTML content
|
| 84 |
-
def format(text_list):
|
| 85 |
format_list = []
|
| 86 |
-
for text in text_list:
|
|
|
|
| 87 |
# Split text by periods
|
| 88 |
sentences = text.split('.')
|
| 89 |
# Create HTML list items
|
| 90 |
list_items = "".join([f"- {sentence.strip()}.\n" for sentence in sentences if sentence.strip()])
|
| 91 |
-
format_list.append(list_items)
|
| 92 |
return format_list
|
| 93 |
|
| 94 |
# Initialize session state for page index and text
|
|
@@ -96,7 +99,7 @@ if (summ_text is not None) or ('summ_text' in st.session_state):
|
|
| 96 |
st.session_state.page_index = 0
|
| 97 |
|
| 98 |
if 'summ_text' not in st.session_state:
|
| 99 |
-
st.session_state.summ_text = format(summ_text)
|
| 100 |
|
| 101 |
if 'current_text' not in st.session_state:
|
| 102 |
st.session_state.current_text = st.session_state.summ_text[st.session_state.page_index]
|
|
|
|
| 36 |
option = st.selectbox(
|
| 37 |
'Select parsing method.',
|
| 38 |
('monkey', 'x2d', 'lxml'))
|
| 39 |
+
|
| 40 |
+
range_values = st.slider(
|
| 41 |
+
'Select a range of values',
|
| 42 |
+
min_value=0,
|
| 43 |
+
max_value=100,
|
| 44 |
+
value=(0, 25)
|
| 45 |
+
)
|
| 46 |
|
| 47 |
summ_text = None
|
| 48 |
|
|
|
|
| 74 |
|
| 75 |
with st.status("Generating slides..."):
|
| 76 |
summarizor = pipeline("summarization", model=summ_model_path, device = device)
|
| 77 |
+
title_list = document.segmentation['key']
|
| 78 |
summ_text = summarizor(document.segmentation['text'], max_length=100, min_length=10, do_sample=False)
|
| 79 |
summ_text = [text['summary_text'] for text in summ_text]
|
| 80 |
|
|
|
|
| 83 |
if (summ_text is not None) or ('summ_text' in st.session_state):
|
| 84 |
|
| 85 |
# Function to render HTML content
|
| 86 |
+
def format(title_list, text_list):
|
| 87 |
format_list = []
|
| 88 |
+
for index, text in enumerate(text_list):
|
| 89 |
+
title = "##" + title_list[index] + "\n"
|
| 90 |
# Split text by periods
|
| 91 |
sentences = text.split('.')
|
| 92 |
# Create HTML list items
|
| 93 |
list_items = "".join([f"- {sentence.strip()}.\n" for sentence in sentences if sentence.strip()])
|
| 94 |
+
format_list.append(title + list_items)
|
| 95 |
return format_list
|
| 96 |
|
| 97 |
# Initialize session state for page index and text
|
|
|
|
| 99 |
st.session_state.page_index = 0
|
| 100 |
|
| 101 |
if 'summ_text' not in st.session_state:
|
| 102 |
+
st.session_state.summ_text = format(summ_text, title_list)
|
| 103 |
|
| 104 |
if 'current_text' not in st.session_state:
|
| 105 |
st.session_state.current_text = st.session_state.summ_text[st.session_state.page_index]
|