Update app.py

app.py

@@ -3,39 +3,48 @@ import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
 import PyPDF2
 
-# Load model and tokenizer
+# ----------------- Model Setup -----------------
 model_name = "ibm-granite/granite-3.2-2b-instruct"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+# Some models require trust_remote_code
+tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
     torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
-    device_map="auto" if torch.cuda.is_available() else None
+    device_map="auto" if torch.cuda.is_available() else None,
+    trust_remote_code=True
 )
 
 # Ensure pad token is set
 if tokenizer.pad_token is None:
     tokenizer.pad_token = tokenizer.eos_token
 
-# ---------- Core Functions ----------
-def generate_response(prompt, max_length=1024):
-    inputs = tokenizer(prompt, return_tensors="pt", truncation=True, max_length=512)
 
-    if torch.cuda.is_available():
-        inputs = {k: v.to(model.device) for k, v in inputs.items()}
-
-    with torch.no_grad():
-        outputs = model.generate(
-            **inputs,
-            max_length=max_length,
-            temperature=0.7,
-            do_sample=True,
-            pad_token_id=tokenizer.eos_token_id
-        )
+# ----------------- Core Functions -----------------
+def generate_response(prompt, max_length=512):
+    try:
+        inputs = tokenizer(prompt, return_tensors="pt", truncation=True, max_length=256)
+
+        if torch.cuda.is_available():
+            inputs = {k: v.to(model.device) for k, v in inputs.items()}
+
+        with torch.no_grad():
+            outputs = model.generate(
+                **inputs,
+                max_length=max_length,
+                temperature=0.7,
+                do_sample=True,
+                pad_token_id=tokenizer.eos_token_id
+            )
+
+        # Slice only the generated continuation
+        response_ids = outputs[0][inputs["input_ids"].shape[-1]:]
+        response = tokenizer.decode(response_ids, skip_special_tokens=True)
+        return response.strip()
+    except Exception as e:
+        return f"⚠️ Error generating response: {str(e)}"
 
-    # Slice out only the generated part
-    response_ids = outputs[0][inputs["input_ids"].shape[-1]:]
-    response = tokenizer.decode(response_ids, skip_special_tokens=True)
-    return response.strip()
 
 def extract_text_from_pdf(pdf_file):
     if pdf_file is None:
@@ -44,70 +53,26 @@ def extract_text_from_pdf(pdf_file):
         pdf_reader = PyPDF2.PdfReader(pdf_file)
         text = ""
         for page in pdf_reader.pages:
-            if page.extract_text():
-                text += page.extract_text() + "\n"
-        return text
+            page_text = page.extract_text()
+            if page_text:
+                text += page_text + "\n"
+        return text.strip()
     except Exception as e:
-        return f"Error reading PDF: {str(e)}"
+        return f"⚠️ Error reading PDF: {str(e)}"
+
 
 def eco_tips_generator(problem_keywords):
+    if not problem_keywords.strip():
+        return "⚠️ Please enter some keywords."
     prompt = (
         f"Generate practical and actionable eco-friendly tips for sustainable living "
         f"related to: {problem_keywords}. Provide specific solutions and suggestions:"
     )
-    return generate_response(prompt, max_length=800)
+    return generate_response(prompt, max_length=400)
+
 
 def policy_summarization(pdf_file, policy_text):
     if pdf_file is not None:
         content = extract_text_from_pdf(pdf_file)
-        summary_prompt = (
-            f"Summarize the following policy document and extract the most important points, "
-            f"key provisions, and implications:\n\n{content}"
-        )
     else:
-        summary_prompt = (
-            f"Summarize the following policy document and extract the most important points, "
-            f"key provisions, and implications:\n\n{policy_text}"
-        )
-
-    return generate_response(summary_prompt, max_length=1200)
-
-# ---------- Gradio Interface ----------
-with gr.Blocks() as app:
-    gr.Markdown("# 🌍 Eco Assistant & Policy Analyzer")
-
-    with gr.Tabs():
-        with gr.TabItem("♻️ Eco Tips Generator"):
-            with gr.Row():
-                with gr.Column():
-                    keywords_input = gr.Textbox(
-                        label="Environmental Problem/Keywords",
-                        placeholder="e.g., plastic, solar, water waste, energy saving...",
-                        lines=3
-                    )
-                    generate_tips_btn = gr.Button("Generate Eco Tips")
-
-                with gr.Column():
-                    tips_output = gr.Textbox(label="Sustainable Living Tips", lines=15)
-
-            generate_tips_btn.click(eco_tips_generator, inputs=keywords_input, outputs=tips_output)
-
-        with gr.TabItem("📑 Policy Summarization"):
-            with gr.Row():
-                with gr.Column():
-                    pdf_upload = gr.File(label="Upload Policy PDF", file_types=[".pdf"])
-                    policy_text_input = gr.Textbox(
-                        label="Or paste policy text here",
-                        placeholder="Paste policy document text...",
-                        lines=5
-                    )
-                    summarize_btn = gr.Button("Summarize Policy")
-
-                with gr.Column():
-                    summary_output = gr.Textbox(label="Policy Summary & Key Points", lines=20)
-
-            summarize_btn.click(policy_summarization, inputs=[pdf_upload, policy_text_input], outputs=summary_output)
-
-# Launch app
-if __name__ == "__main__":
-    app.launch(share=True)
+        content = polic