add LLM for recommendation

app.py

@@ -4,9 +4,11 @@ import numpy as np
 from PIL import Image
 import io
 from fastapi.middleware.cors import CORSMiddleware
-
+from llm_client import LLMClient
+import json
 # Initialize FastAPI app
 app = FastAPI(title="Image Classification API")
+import logging
 
 # Add CORS middleware
 app.add_middleware(
@@ -17,6 +19,8 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
+
+
 # Load the Keras model once at startup
 try:
     model = load_model('IAPLD.h5')
@@ -44,6 +48,34 @@ def preprocess_image(image: Image.Image) -> np.ndarray:
 @app.get("/")
 async def root():
     return {"message": "Welcome to the Image Classification API. Use POST /predict/ to upload an image."}
+from recomm import Redommend
+
+@app.get("/recommendation")
+async def recommendation(disease: str):
+    if not disease:
+        raise HTTPException(status_code=400, detail="Disease parameter is required")
+
+    try:
+        llm_client = LLMClient()
+        recommender = Redommend(llm_client)
+
+        raw_response = recommender._run(disease).strip()
+
+        if raw_response.startswith("```json"):
+            raw_response = raw_response.replace("```json", "").replace("```", "").strip()
+
+        data = json.loads(raw_response)
+        return data
+
+    except json.JSONDecodeError:
+        raise HTTPException(
+            status_code=500,
+            detail=f"Le LLM n’a pas renvoyé un JSON valide : {raw_response}"
+
+        )
+    except Exception as e:
+        logging.error(f"Error in recommendation endpoint: {str(e)}")
+        raise HTTPException(status_code=500, detail=f"Error processing request: {str(e)}")
 
 # Prediction endpoint
 @app.post("/predict/")

llm_client.py

@@ -0,0 +1,13 @@
+# app/llm_client.py
+from langchain_google_genai import ChatGoogleGenerativeAI
+
+class LLMClient:
+    def __init__(self, model: str = "models/gemini-2.0-flash", temp: float = 0.5):
+        self.llm = ChatGoogleGenerativeAI(
+            model=model,
+            google_api_key= "AIzaSyAmpT2kjqFcz7HZyiFeh6dBOu-zx_MRxIA",
+            temperature=temp
+        )
+
+    def invoke(self, prompt: str) -> str:
+        return self.llm.invoke(prompt).content

recomm.py

@@ -0,0 +1,37 @@
+from langchain.prompts import PromptTemplate
+from llm_client import LLMClient
+from langchain.agents import Tool
+
+class Redommend:
+    ## class for generating recommendations based on user input
+    def __init__(self, llm_client: LLMClient):
+        self.llm_client = llm_client
+
+        # Define the prompt template for generating recommendations.
+        # Note: now the input variable is called "maladie" (same as the placeholder).
+        template = """
+        Vous êtes en culture de pommes de terre 
+        * Pour guérir cette  {maladie} : 
+        - Vous devez recommander les meilleures pratiques de culture pour cette maladie.
+        - Veuillez lister les recommandations de manière concise, étape par étape.
+        - Donnez la réponse sous forme de liste numérotée, en JSON.
+        - Donnez simplement la réponse en JSON, sans explications supplémentaires.
+        - Donnez la reponse  en Français.
+        """
+        self.prompt_template = PromptTemplate(
+            input_variables=["maladie"],
+            template=template
+        )
+
+        self.tool = Tool(
+            name="recommender",
+            func=self._run,
+            description="Use this tool to get recommendations for potato cultivation based on the disease."
+        )
+
+    def _run(self, maladie: str) -> str:
+        # Generate the prompt with the maladie string that comes from the query parameter.
+        prompt = self.prompt_template.format(maladie=maladie)
+        # Invoke the LLM client to get recommendations
+        response = self.llm_client.invoke(prompt)
+        return response

requirements.txt

@@ -2,4 +2,7 @@ fastapi==0.115.4
 uvicorn==0.32.0
 tensorflow==2.17.1
 numpy==1.26.4
-pillow==10.2.0
+pillow==10.2.0
+langchain==0.2.0
+langchain-google-genai
+accelerate>=0.26.0

schemas.py

@@ -0,0 +1,9 @@
+from pydantic import BaseModel
+from typing import List
+
+class GeneralResponse(BaseModel):
+    status: str
+    message: str
+    data : List[str]
+class Request(BaseModel):
+    disease: str