update all

app.py

@@ -14,14 +14,14 @@ tokenizer = BertTokenizerFast.from_pretrained(tokenizer_save_path)
 
 sentiment_mapping = {0: 'neutral', 1: 'negative', 2: 'positive'}
 
-# Function to predict the sentiment of the text
+# function to predict the sentiment of the text
 def predict_sentiment(text):
     inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=512).to(device)
     outputs = model(**inputs)
     predictions = torch.argmax(outputs.logits, dim=-1).cpu().numpy()
     return sentiment_mapping[predictions[0]]
 
-# Function to scrape headlines and predict their sentiment
+# function to scrape headlines and predict their sentiment
 def scrape_and_predict():
     url = "https://finance.yahoo.com"
     response = requests.get(url)
@@ -37,15 +37,13 @@ def scrape_and_predict():
 
     return headlines
 
-# Gradio interface for sentiment prediction
 def sentiment_interface(text):
     return predict_sentiment(text)
 
-# Gradio interface for scraping headlines and predicting their sentiment
 def scrape_interface():
     return scrape_and_predict()
 
-# Define Gradio blocks for the app
+# gradio app
 with gr.Blocks() as demo:
     gr.Markdown("# Sentiment Analysis and News Scraping")
     
@@ -60,6 +58,5 @@ with gr.Blocks() as demo:
         headlines_output = gr.JSON(label="Headlines and Sentiment")
         scrape_button.click(scrape_interface, outputs=headlines_output)
 
-# Launch the Gradio app
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch(share=True)

requirements.txt

@@ -5,4 +5,5 @@ transformers==4.31.0
 beautifulsoup4==4.12.2
 requests==2.31.0
 gradio==3.23.0
-safetensors==0.3.2
+safetensors==0.3.2
+numpy<2