Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -3,10 +3,6 @@ import json
|
|
| 3 |
import pandas as pd
|
| 4 |
import streamlit.components.v1 as components
|
| 5 |
|
| 6 |
-
# Initialize session state for tracking the last clicked row
|
| 7 |
-
if 'last_clicked_row' not in st.session_state:
|
| 8 |
-
st.session_state['last_clicked_row'] = None
|
| 9 |
-
|
| 10 |
# Function to load JSONL file into a DataFrame
|
| 11 |
def load_jsonl(file_path):
|
| 12 |
data = []
|
|
@@ -20,7 +16,8 @@ def filter_by_keyword(df, keyword):
|
|
| 20 |
return df[df.apply(lambda row: row.astype(str).str.contains(keyword).any(), axis=1)]
|
| 21 |
|
| 22 |
# Function to generate HTML with textarea
|
| 23 |
-
def generate_html_with_textarea(
|
|
|
|
| 24 |
return f'''
|
| 25 |
<!DOCTYPE html>
|
| 26 |
<html>
|
|
@@ -37,7 +34,7 @@ def generate_html_with_textarea(text_to_speak):
|
|
| 37 |
<body>
|
| 38 |
<h1>🔊 Read It Aloud</h1>
|
| 39 |
<textarea id="textArea" rows="10" cols="80">
|
| 40 |
-
{
|
| 41 |
</textarea>
|
| 42 |
<br>
|
| 43 |
<button onclick="readAloud()">🔊 Read Aloud</button>
|
|
@@ -61,6 +58,16 @@ data = large_data if file_option == "usmle_16.2MB.jsonl" else small_data
|
|
| 61 |
# Top 20 healthcare terms for USMLE
|
| 62 |
top_20_terms = ['Heart', 'Lung', 'Pain', 'Memory', 'Kidney', 'Diabetes', 'Cancer', 'Infection', 'Virus', 'Bacteria', 'Gastrointestinal', 'Skin', 'Blood', 'Surgery']
|
| 63 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
# Create Expander and Columns UI for terms
|
| 65 |
with st.expander("Search by Common Terms 📚"):
|
| 66 |
cols = st.columns(4)
|
|
@@ -69,31 +76,19 @@ with st.expander("Search by Common Terms 📚"):
|
|
| 69 |
if st.button(f"{term}"):
|
| 70 |
filtered_data = filter_by_keyword(data, term)
|
| 71 |
st.write(f"Filter on '{term}' 📊")
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
|
| 78 |
-
documentHTML5 = generate_html_with_textarea(question_text)
|
| 79 |
-
html_blocks.append(documentHTML5)
|
| 80 |
-
all_html = ''.join(html_blocks)
|
| 81 |
-
components.html(all_html, width=1280, height=1024)
|
| 82 |
|
| 83 |
# Text input for search keyword
|
| 84 |
search_keyword = st.text_input("Or, enter a keyword to filter data:")
|
| 85 |
if st.button("Search 🕵️♀️"):
|
| 86 |
filtered_data = filter_by_keyword(data, search_keyword)
|
| 87 |
st.write(f"Filtered Dataset by '{search_keyword}' 📊")
|
| 88 |
-
|
| 89 |
-
if not filtered_data.empty:
|
| 90 |
-
html_blocks = []
|
| 91 |
-
for idx, row in filtered_data.iterrows():
|
| 92 |
-
question_text = row.get("question", "No question field")
|
| 93 |
-
documentHTML5 = generate_html_with_textarea(question_text)
|
| 94 |
-
html_blocks.append(documentHTML5)
|
| 95 |
-
all_html = ''.join(html_blocks)
|
| 96 |
-
components.html(all_html, width=1280, height=1024)
|
| 97 |
|
| 98 |
|
| 99 |
|
|
|
|
| 3 |
import pandas as pd
|
| 4 |
import streamlit.components.v1 as components
|
| 5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
# Function to load JSONL file into a DataFrame
|
| 7 |
def load_jsonl(file_path):
|
| 8 |
data = []
|
|
|
|
| 16 |
return df[df.apply(lambda row: row.astype(str).str.contains(keyword).any(), axis=1)]
|
| 17 |
|
| 18 |
# Function to generate HTML with textarea
|
| 19 |
+
def generate_html_with_textarea(row):
|
| 20 |
+
first_three_columns_text = ' '.join([f"{col}: {row[col]}" for col in row.index[:3]])
|
| 21 |
return f'''
|
| 22 |
<!DOCTYPE html>
|
| 23 |
<html>
|
|
|
|
| 34 |
<body>
|
| 35 |
<h1>🔊 Read It Aloud</h1>
|
| 36 |
<textarea id="textArea" rows="10" cols="80">
|
| 37 |
+
{first_three_columns_text}
|
| 38 |
</textarea>
|
| 39 |
<br>
|
| 40 |
<button onclick="readAloud()">🔊 Read Aloud</button>
|
|
|
|
| 58 |
# Top 20 healthcare terms for USMLE
|
| 59 |
top_20_terms = ['Heart', 'Lung', 'Pain', 'Memory', 'Kidney', 'Diabetes', 'Cancer', 'Infection', 'Virus', 'Bacteria', 'Gastrointestinal', 'Skin', 'Blood', 'Surgery']
|
| 60 |
|
| 61 |
+
# Initialize session state for tracking the clicked row in DataFrame
|
| 62 |
+
if 'clicked_row' not in st.session_state:
|
| 63 |
+
st.session_state['clicked_row'] = None
|
| 64 |
+
|
| 65 |
+
# Function to display the DataFrame and capture clicks
|
| 66 |
+
def display_clickable_dataframe(df):
|
| 67 |
+
for idx, row in df.iterrows():
|
| 68 |
+
if st.button(f"Row {idx}", key=f"row_{idx}"):
|
| 69 |
+
st.session_state['clicked_row'] = row
|
| 70 |
+
|
| 71 |
# Create Expander and Columns UI for terms
|
| 72 |
with st.expander("Search by Common Terms 📚"):
|
| 73 |
cols = st.columns(4)
|
|
|
|
| 76 |
if st.button(f"{term}"):
|
| 77 |
filtered_data = filter_by_keyword(data, term)
|
| 78 |
st.write(f"Filter on '{term}' 📊")
|
| 79 |
+
display_clickable_dataframe(filtered_data)
|
| 80 |
+
|
| 81 |
+
# Display the HTML content for the clicked row
|
| 82 |
+
if st.session_state['clicked_row'] is not None:
|
| 83 |
+
html_content = generate_html_with_textarea(st.session_state['clicked_row'])
|
| 84 |
+
components.html(html_content, width=1280, height=1024)
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
|
| 86 |
# Text input for search keyword
|
| 87 |
search_keyword = st.text_input("Or, enter a keyword to filter data:")
|
| 88 |
if st.button("Search 🕵️♀️"):
|
| 89 |
filtered_data = filter_by_keyword(data, search_keyword)
|
| 90 |
st.write(f"Filtered Dataset by '{search_keyword}' 📊")
|
| 91 |
+
display_clickable_dataframe(filtered_data)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 92 |
|
| 93 |
|
| 94 |
|