Update app.py

app.py

@@ -1,92 +1,93 @@
-# %%
-import torch
-import numpy as np
-import pandas as pd
-from datetime import datetime
-from newsapi import NewsApiClient
-from transformers import pipeline
-import plotly.graph_objects as go
-import gradio as gr
-
-# Initialize News API client with your API key
-newsapi = NewsApiClient(api_key='381793b3d6834758918838bca0cf52ee')
-
-# Define sentiment analyzer using FinBERT
-sentiment_analyzer = pipeline("text-classification", model="ProsusAI/finbert")
-
-# Function to fetch news, analyze sentiment, and create an interactive Plotly plot
-def analyze_news_sentiment(company_name):
-    # Fetch news articles related to the company
-    news = newsapi.get_everything(q=company_name, language='en', sort_by='publishedAt')
-    
-    # Extract headlines from news articles
-    headlines = [article['title'] for article in news['articles']]
-    
-    # Perform sentiment analysis on the headlines
-    result = sentiment_analyzer(headlines)
-    df = pd.DataFrame(result)
-    
-    # Map labels to numeric values
-    label_mapping = {'positive': 1, 'neutral': 0, 'negative': -1}
-    df['sentiment'] = df['label'].map(label_mapping)
-    
-    # Drop the 'label' column
-    df.drop(columns=['label'], inplace=True)
-    
-    # Filter out neutral sentiment values
-    positive_sentiment = df[df['sentiment'] == 1]['sentiment']
-    negative_sentiment = df[df['sentiment'] == -1]['sentiment']
-
-    # Create the interactive Plotly histogram
-    fig = go.Figure()
-
-    # Add Positive sentiment histogram
-    fig.add_trace(go.Histogram(
-        x=positive_sentiment,
-        nbinsx=1,
-        name='Positive',
-        marker_color='purple',
-        opacity=0.75
-    ))
-
-    # Add Negative sentiment histogram
-    fig.add_trace(go.Histogram(
-        x=negative_sentiment,
-        nbinsx=1,
-        name='Negative',
-        marker_color='skyblue',
-        opacity=0.75
-    ))
-
-    # Update layout for better visualization
-    fig.update_layout(
-        title=f'Sentiment Distribution for {company_name}',
-        xaxis_title='Sentiment',
-        yaxis_title='Count',
-        barmode='overlay',
-        plot_bgcolor='black',
-        paper_bgcolor='black',
-        font=dict(color='white'),
-        xaxis=dict(tickvals=[-1, 1], ticktext=['Negative', 'Positive']),
-        bargap=0.2,
-    )
-
-    return fig
-
-
-
-
-# %%
-# Create a Gradio interface
-interface = gr.Interface(
-    fn=analyze_news_sentiment,  # Function to run
-    inputs=gr.Textbox(label="Enter Company Name"),  # Input: company name
-    outputs=gr.Plot(label="Sentiment Distribution"),  # Output: Interactive Plotly chart
-    title="Sentiment Analysis on News Headlines",
-    description="Enter a company name to analyze the sentiment of the latest news related to that company."
-)
-
-# Launch the Gradio app
-interface.launch()
-
-
+# %%
+import torch
+import numpy as np
+import pandas as pd
+from datetime import datetime
+from newsapi import NewsApiClient
+from transformers import pipeline
+import plotly.graph_objects as go
+import gradio as gr
+
+# Initialize News API client with your API key
+newsapi = NewsApiClient(api_key='381793b3d6834758918838bca0cf52ee')
+
+# Define sentiment analyzer using FinBERT
+sentiment_analyzer = pipeline("text-classification", model="ProsusAI/finbert")
+
+# Function to fetch news, analyze sentiment, and create an interactive Plotly plot
+def analyze_news_sentiment(company_name):
+    # Fetch news articles related to the company
+    news = newsapi.get_everything(q=company_name, language='en', sort_by='publishedAt')
+    
+    # Extract headlines from news articles
+    headlines = [article['title'] for article in news['articles']]
+    
+    # Perform sentiment analysis on the headlines
+    result = sentiment_analyzer(headlines)
+    df = pd.DataFrame(result)
+    
+    # Map labels to numeric values
+    label_mapping = {'positive': 1, 'neutral': 0, 'negative': -1}
+    df['sentiment'] = df['label'].map(label_mapping)
+    
+    # Drop the 'label' column
+    df.drop(columns=['label'], inplace=True)
+    
+    # Filter out neutral sentiment values
+    positive_sentiment = df[df['sentiment'] == 1]['sentiment']
+    negative_sentiment = df[df['sentiment'] == -1]['sentiment']
+
+    # Create the interactive Plotly histogram
+    fig = go.Figure()
+
+    # Add Positive sentiment histogram
+    fig.add_trace(go.Histogram(
+        x=positive_sentiment,
+        nbinsx=1,
+        name='Positive',
+        marker_color='purple',
+        opacity=0.75
+    ))
+
+    # Add Negative sentiment histogram
+    fig.add_trace(go.Histogram(
+        x=negative_sentiment,
+        nbinsx=1,
+        name='Negative',
+        marker_color='skyblue',
+        opacity=0.75
+    ))
+
+    # Update layout for better visualization
+    fig.update_layout(
+        title=f'Sentiment Distribution for {company_name}',
+        xaxis_title='Sentiment',
+        yaxis_title='Count',
+        barmode='overlay',
+        plot_bgcolor='black',
+        paper_bgcolor='black',
+        font=dict(color='white'),
+        xaxis=dict(tickvals=[-1, 1], ticktext=['Negative', 'Positive']),
+        bargap=0.2,
+    )
+
+    return fig
+
+
+
+
+# %%
+# Create a Gradio interface
+interface = gr.Interface(
+    fn=analyze_news_sentiment,  # Function to run
+    inputs=gr.Textbox(label="Enter Company Name"),  # Input: company name
+    outputs=gr.Plot(label="Sentiment Distribution"),  # Output: Interactive Plotly chart
+    title="Sentiment Analysis on News Headlines",
+    description="Enter a company name to analyze the sentiment of the latest news related to that company.",
+    theme="dark"  # Optional theme setting for Gradio 
+)
+
+# Launch the Gradio app
+interface.launch()
+
+