Update app.py

app.py

@@ -1,22 +1,34 @@
-from fastapi import FastAPI
-from pydantic import BaseModel
-from transformers import pipeline
+from flask import Flask, request, jsonify
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
 
-app = FastAPI()
+app = Flask(__name__)
 
-# Charger le modèle Hugging Face
-model_name = "fatmata/psybot"
-generator = pipeline("text-generation", model=model_name)
+# Charger le modèle depuis Hugging Face
+MODEL_NAME = "fatmata/psybot"  # Remplace avec le vrai nom de ton modèle
+tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, torch_dtype=torch.float16)
 
-# Définition du format des requêtes
-class TextRequest(BaseModel):
-    prompt: str
+@app.route("/chat", methods=["POST"])
+def chat():
+    data = request.json
+    user_input = data.get("message", "")
 
-@app.get("/")
-def home():
-    return {"message": "Bienvenue sur l'API PsyBot !"}
+    if not user_input:
+        return jsonify({"error": "Message vide"}), 400
 
-@app.post("/generate/")
-def generate_text(request: TextRequest):
-    response = generator(request.prompt, max_length=100, do_sample=True)
-    return {"response": response[0]["generated_text"]}
+    # Génération de la réponse
+    prompt = f"<|startoftext|><|user|> {user_input} <|bot|>"
+    inputs = tokenizer(prompt, return_tensors="pt").input_ids.to(model.device)
+
+    with torch.no_grad():
+        output = model.generate(inputs, max_new_tokens=100, pad_token_id=tokenizer.eos_token_id)
+
+    response = tokenizer.decode(output[0], skip_special_tokens=True)
+    if "<|bot|>" in response:
+        response = response.split("<|bot|>")[-1].strip()
+
+    return jsonify({"response": response})
+
+if __name__ == "__main__":
+    app.run(host="0.0.0.0", port=7860)