Update pipeline.py

pipeline.py

@@ -8,7 +8,7 @@ from langchain.embeddings import HuggingFaceEmbeddings
 from langchain.vectorstores import FAISS
 from langchain.chains import RetrievalQA
 from smolagents import CodeAgent, DuckDuckGoSearchTool, ManagedAgent, LiteLLMModel
-# import litellm
+import subprocess  # Import subprocess to run shell commands
 
 # Import the chain builders from our separate files
 from classification_chain import get_classification_chain
@@ -22,8 +22,21 @@ if not os.environ.get("GEMINI_API_KEY"):
 if not os.environ.get("GROQ_API_KEY"):
     os.environ["GROQ_API_KEY"] = getpass.getpass("Enter your GROQ API Key: ")
 
-# 2) Load spaCy model for NER
-nlp = spacy.load("en_core_web_sm")
+# 2) Load spaCy model for NER and download the spaCy model if not already installed
+def install_spacy_model():
+    try:
+        # Check if the model is already installed
+        spacy.load("en_core_web_sm")
+        print("spaCy model 'en_core_web_sm' is already installed.")
+    except OSError:
+        # If model is not installed, download it using subprocess
+        print("Downloading spaCy model 'en_core_web_sm'...")
+        subprocess.run(["python", "-m", "spacy", "download", "en_core_web_sm"], check=True)
+        print("spaCy model 'en_core_web_sm' downloaded successfully.")
+
+# Call the function to install the spaCy model if needed
+install_spacy_model()
+
 
 # Function to extract the main topic using NER
 def extract_main_topic(query: str) -> str: