Made changes to the app, requirements, runtime and added Procfile

Procfile

@@ -0,0 +1 @@
+web: uvicorn app:app --host 0.0.0.0 --port $PORT

app/app.py

@@ -8,6 +8,7 @@ from PyPDF2 import PdfReader
 from sklearn.metrics.pairwise import cosine_similarity
 import torch
 import os
+import numpy as np
 import shutil
 import uuid
 import tempfile
@@ -54,11 +55,6 @@ def load_models_impl():
     model_status["last_error"] = None
 
     try:
-        # Placeholder for the code that should be inside the try block
-        pass
-    except Exception as e:
-        logger.error(f"An error occurred: {e}")
-        raise HTTPException(status_code=500, detail="An internal error occurred.")
         # Check Hugging Face Hub connectivity
         response = requests.head("https://huggingface.co", timeout=5)
         if response.status_code == 200:
@@ -271,15 +267,53 @@ async def analyze_essay(file: UploadFile = File(...), background_tasks: Backgrou
             # Run internal plagiarism detection
             chunks = chunk_text(essay_text)
             if len(chunks) < 2:
-                raise HTTPException(status_code=400, detail="Not enough text chunks to assess internal plagiarism.")
+                return JSONResponse(content={
+                    "ai_detection": ai_result,
+                    "internal_plagiarism": {
+                        "detected": False,
+                        "message": "Not enough text chunks to assess internal plagiarism."
+                    }
+                })
 
             embeddings = embedder.encode(chunks)
-            similarities = []
-        except Exception as e:
-            logger.error(f"An error occurred during analysis: {e}")
-            raise HTTPException(status_code=500, detail="An error occurred during analysis.")
+            
+            # Calculate similarity matrix (excluding self-comparisons)
+            similarity_matrix = cosine_similarity(embeddings)
+            np.fill_diagonal(similarity_matrix, 0)  # Set diagonal to 0 to ignore self-similarity
+            
+            # Find highest similarity pairs
+            plagiarism_threshold = 0.85  # Adjust as needed
+            high_similarities = []
+            
+            for i in range(len(chunks)):
+                for j in range(i+1, len(chunks)):
+                    similarity = similarity_matrix[i][j]
+                    if similarity > plagiarism_threshold:
+                        high_similarities.append({
+                            "chunk1_index": i,
+                            "chunk2_index": j,
+                            "chunk1_text": chunks[i],
+                            "chunk2_text": chunks[j],
+                            "similarity": round(float(similarity), 2)
+                        })
+            
+            # Sort by similarity (highest first)
+            high_similarities.sort(key=lambda x: x["similarity"], reverse=True)
+            
+            return JSONResponse(content={
+                "ai_detection": ai_result,
+                "internal_plagiarism": {
+                    "detected": len(high_similarities) > 0,
+                    "similarity_pairs": high_similarities[:5],  # Return top 5 similar pairs
+                    "total_similar_pairs": len(high_similarities)
+                }
+            })
+            
         except Exception as e:
             logger.error(f"An error occurred during analysis: {e}")
-            raise HTTPException(status_code=500, detail="An error occurred during analysis.")
-           
- 
+            raise HTTPException(status_code=500, detail=f"An error occurred during analysis: {str(e)}")
+
+@app.get("/")
+async def root():
+    """Root endpoint that redirects to docs"""
+    return {"message": "Essay Grader API is running. Visit /docs for the API documentation."}

requirements.txt

@@ -1,15 +1,11 @@
-fastapi==0.115.12
+fastapi==0.115.0
 uvicorn==0.34.0
-transformers==4.50.2
-torch==2.6.0
-sentence-transformers==4.1.0
-datasets
-evaluate
-huggingface-hub
-scikit-learn
-pandas
-numpy
-aiohttp
-python-multipart
-python-dotenv
-requests
+transformers==4.39.3
+sentence-transformers==2.5.1
+torch==2.2.2
+scikit-learn==1.4.0
+PyPDF2==3.0.1
+numpy==1.26.4
+pandas==2.2.1
+requests==2.31.0
+python-multipart==0.0.9

runtime.txt

@@ -1 +1 @@
-python-3.10
+python-3.11.9