Update app.py

app.py

@@ -1,71 +1,52 @@
-from googlesearch import search
+import gradio as gr
+from transformers import pipeline
 from bs4 import BeautifulSoup
 import requests
-from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
 
-# Initialize Phi-3-vision-128k-instruct tokenizer and model
-tokenizer = AutoTokenizer.from_pretrained('microsoft/Phi-3-vision-128k-instruct')
-model = AutoModelForSeq2SeqLM.from_pretrained('microsoft/Phi-3-vision-128k-instruct', trust_remote_code=True)
+def summarize_blog_post(url):
+    # Load summarization pipeline
+    summarizer = pipeline("summarization")
 
-# Function to perform Google search and retrieve URLs, filtering by domain
-def google_search(query: str, num_results: int = 10):
-    """Perform a Google search and retrieve the URLs of the search results."""
-    search_results = []
-    try:
-        for url in search(query, num_results=num_results, domains=["tesla.com", "cnbc.com", "reuters.com", "bloomberg.com", "investopedia.com"]):
-            search_results.append(url)
-    except TypeError:
-        for url in search(query, num_results=num_results):
-            if any(domain in url for domain in ["tesla.com", "cnbc.com", "reuters.com", "bloomberg.com", "investopedia.com"]):
-                search_results.append(url)
-    return search_results
+    # Get blog post content
+    r = requests.get(url)
+    soup = BeautifulSoup(r.text, 'html.parser')
+    results = soup.find_all(['h1', 'p'])
+    text = [result.text for result in results]
+    ARTICLE = ' '.join(text)
 
-# Function to fetch content from a URL and summarize it
-def fetch_and_summarize_url(url: str):
-    try:
-        response = requests.get(url)
-        response.raise_for_status()
-        soup = BeautifulSoup(response.text, 'html.parser')
-        
-        # Extract relevant content (e.g., paragraphs or sections)
-        paragraphs = [p.text for p in soup.find_all('p')]
-        combined_text = " ".join(paragraphs[:3])  # Combine first few paragraphs for summary
-        
-        # Tokenize the text
-        inputs = tokenizer.encode("summarize: " + combined_text, return_tensors="pt", max_length=1024, truncation=True)
-        
-        # Generate summary
-        summary_ids = model.generate(inputs, max_length=150, min_length=40, length_penalty=2.0, num_beams=4, early_stopping=True)
-        summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
-        
-        return summary
-    except requests.RequestException as e:
-        return None
+    # Chunk text
+    max_chunk = 500
+    ARTICLE = ARTICLE.replace('.', '.<eos>')
+    ARTICLE = ARTICLE.replace('?', '?<eos>')
+    ARTICLE = ARTICLE.replace('!', '!<eos>')
 
-# Function to perform Google search and aggregate summaries
-def google_search_and_answer(question: str, keywords: str):
-    search_query = f"{question} {keywords}"
-    search_results = google_search(search_query)
-    summaries = []
-    for url in search_results:
-        fetched_summary = fetch_and_summarize_url(url)
-        if fetched_summary:
-            # Add additional logic to filter summaries based on relevance
-            # Example: Check if either question or keywords are present in fetched_summary
-            if question.lower() in fetched_summary.lower() or keywords.lower() in fetched_summary.lower():
-                summaries.append(fetched_summary)
-    if summaries:
-        return "\n\n".join(summaries)
-    else:
-        return "No relevant information found."
+    sentences = ARTICLE.split('<eos>')
+    current_chunk = 0
+    chunks = []
+    for sentence in sentences:
+        if len(chunks) == current_chunk + 1:
+            if len(chunks[current_chunk]) + len(sentence.split(' ')) <= max_chunk:
+                chunks[current_chunk].extend(sentence.split(' '))
+            else:
+                current_chunk += 1
+                chunks.append(sentence.split(' '))
+        else:
+            chunks.append(sentence.split(' '))
 
-# Main function to run the script
-def main():
-    print("Intelligent Assistant")
-    question = input("Enter your query: ")
-    keywords = input("Enter specific keywords (e.g., 'Q1 2024 financial results Tesla'): ")
-    answer = google_search_and_answer(question, keywords)
-    print("Answer:", answer)
+    for chunk_id in range(len(chunks)):
+        chunks[chunk_id] = ' '.join(chunks[chunk_id])
 
-if __name__ == "__main__":
-    main()
+    # Summarize text
+    summaries = summarizer(chunks, max_length=120, min_length=30, do_sample=False)
+    summary_text = " ".join([summary['summary_text'] for summary in summaries])
+    return summary_text
+
+iface = gr.Interface(
+    fn=summarize_blog_post, 
+    inputs="text", 
+    outputs="text",
+    title="Medium Blog Post Summarizer",
+    description="Enter the URL of a Medium blog post to get a summarized version of the content."
+)
+
+iface.launch()