Create utils.py

utils.py

@@ -0,0 +1,76 @@
+import fitz  # PyMuPDF
+import re
+import faiss
+import numpy as np
+from sentence_transformers import SentenceTransformer
+import requests
+
+# Load embedding model
+embed_model = SentenceTransformer("all-MiniLM-L6-v2")
+
+def extract_pdf_by_page(path):
+    doc = fitz.open(path)
+    pages = []
+    for page_num, page in enumerate(doc, 1):
+        text = page.get_text()
+        pages.append({"page_number": page_num, "text": text})
+    return pages
+
+def split_by_chapter(pages):
+    chapters = []
+    current = {"title": "Introduction", "text": "", "pages": []}
+    chapter_pattern = re.compile(r'chapter\s+\d+[:.\s]', re.IGNORECASE)
+
+    for page in pages:
+        if chapter_pattern.search(page["text"]):
+            chapters.append(current)
+            current = {
+                "title": chapter_pattern.search(page["text"]).group().strip(),
+                "text": page["text"],
+                "pages": [page["page_number"]],
+            }
+        else:
+            current["text"] += "\n" + page["text"]
+            current["pages"].append(page["page_number"])
+    chapters.append(current)
+    return chapters
+
+def build_faiss_index(chunks):
+    texts = [chunk["text"] for chunk in chunks]
+    embeddings = embed_model.encode(texts, convert_to_numpy=True)
+    index = faiss.IndexFlatL2(embeddings.shape[1])
+    index.add(embeddings)
+    return index, texts, chunks
+
+def retrieve_text(query, index, texts, chunks, top_k=1):
+    query_vec = embed_model.encode([query])
+    D, I = index.search(query_vec, top_k)
+    return [chunks[i] for i in I[0]]
+
+def generate_notes_questions(text, groq_api_key, model="meta-llama/llama-4-scout-17b-16e-instruct"):
+    prompt = f"""
+You are an educational assistant. Given the following content, generate:
+1. Bullet point summary notes.
+2. Five comprehension questions with their answers.
+
+--- BEGIN CONTENT ---
+{text[:3000]}
+--- END CONTENT ---
+"""
+
+    headers = {
+        "Authorization": f"Bearer {groq_api_key}",
+        "Content-Type": "application/json"
+    }
+    payload = {
+        "model": model,
+        "messages": [
+            {"role": "user", "content": prompt}
+        ],
+        "temperature": 0.7
+    }
+
+    url = "https://api.groq.com/openai/v1/chat/completions"
+    response = requests.post(url, headers=headers, json=payload)
+    response.raise_for_status()
+    return response.json()['choices'][0]['message']['content']