removed api key

generate_index.py

@@ -1,67 +1,67 @@
-import os
-import pdfplumber
-import pickle
-import faiss
-import numpy as np
-from langchain_google_genai import GoogleGenerativeAIEmbeddings
-from langchain.vectorstores import FAISS
-
-# Configuration
-TEMPLATE_DIR = "dataset"  # Folder containing template answer PDFs
-INDEX_NAME = "index"       # Prefix for FAISS index files
-API_KEY = "AIzaSyArdn9_Uabo9q0aYmm4dxybVEb0tj7dlrk"
-
-def extract_text_from_pdf(pdf_path):
-    """Extracts text from a single PDF file."""
-    text = ""
-    with pdfplumber.open(pdf_path) as pdf_reader:
-        for page in pdf_reader.pages:
-            text += page.extract_text() or ""  # Handle NoneType
-    return text.strip()
-
-def process_template_answers():
-    """Extracts answers from template PDFs and stores them in FAISS."""
-    template_answers = {}
-    
-    for file in os.listdir(TEMPLATE_DIR):
-        if file.endswith(".pdf"):
-            question_number = file.replace(".pdf", "").upper()  # Extract question ID (e.g., 1A)
-            file_path = os.path.join(TEMPLATE_DIR, file)
-            extracted_text = extract_text_from_pdf(file_path)
-            if extracted_text:
-                template_answers[question_number] = extracted_text
-    
-    return template_answers
-
-def generate_faiss_index(api_key):
-    """Creates FAISS index with Google AI Embeddings."""
-    print("🔄 Extracting template answers...")
-    template_answers = process_template_answers()
-    
-    if not template_answers:
-        print("❌ No valid template answers found.")
-        return
-
-    print("🔍 Generating embeddings...")
-    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001", google_api_key=api_key)
-    
-    texts = list(template_answers.values())
-    question_numbers = list(template_answers.keys())
-
-    text_embeddings = np.array([embeddings.embed_query(text) for text in texts]).astype('float32')
-
-    print("📁 Creating FAISS index...")
-    dimension = text_embeddings.shape[1]
-    index = faiss.IndexFlatL2(dimension)
-    index.add(text_embeddings)
-
-    print("💾 Saving FAISS index...")
-    faiss.write_index(index, f"{INDEX_NAME}.faiss")
-
-    with open(f"{INDEX_NAME}.pkl", "wb") as f:
-        pickle.dump(question_numbers, f)
-
-    print("✅ Indexing complete!")
-
-if __name__ == "__main__":
+import os
+import pdfplumber
+import pickle
+import faiss
+import numpy as np
+from langchain_google_genai import GoogleGenerativeAIEmbeddings
+from langchain.vectorstores import FAISS
+
+# Configuration
+TEMPLATE_DIR = "dataset"  # Folder containing template answer PDFs
+INDEX_NAME = "index"       # Prefix for FAISS index files
+API_KEY = "" # add google api key
+
+def extract_text_from_pdf(pdf_path):
+    """Extracts text from a single PDF file."""
+    text = ""
+    with pdfplumber.open(pdf_path) as pdf_reader:
+        for page in pdf_reader.pages:
+            text += page.extract_text() or ""  # Handle NoneType
+    return text.strip()
+
+def process_template_answers():
+    """Extracts answers from template PDFs and stores them in FAISS."""
+    template_answers = {}
+    
+    for file in os.listdir(TEMPLATE_DIR):
+        if file.endswith(".pdf"):
+            question_number = file.replace(".pdf", "").upper()  # Extract question ID (e.g., 1A)
+            file_path = os.path.join(TEMPLATE_DIR, file)
+            extracted_text = extract_text_from_pdf(file_path)
+            if extracted_text:
+                template_answers[question_number] = extracted_text
+    
+    return template_answers
+
+def generate_faiss_index(api_key):
+    """Creates FAISS index with Google AI Embeddings."""
+    print("🔄 Extracting template answers...")
+    template_answers = process_template_answers()
+    
+    if not template_answers:
+        print("❌ No valid template answers found.")
+        return
+
+    print("🔍 Generating embeddings...")
+    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001", google_api_key=api_key)
+    
+    texts = list(template_answers.values())
+    question_numbers = list(template_answers.keys())
+
+    text_embeddings = np.array([embeddings.embed_query(text) for text in texts]).astype('float32')
+
+    print("📁 Creating FAISS index...")
+    dimension = text_embeddings.shape[1]
+    index = faiss.IndexFlatL2(dimension)
+    index.add(text_embeddings)
+
+    print("💾 Saving FAISS index...")
+    faiss.write_index(index, f"{INDEX_NAME}.faiss")
+
+    with open(f"{INDEX_NAME}.pkl", "wb") as f:
+        pickle.dump(question_numbers, f)
+
+    print("✅ Indexing complete!")
+
+if __name__ == "__main__":
     generate_faiss_index(API_KEY)