delete test_api_genai

Tests_API_GenAI.py

@@ -1,130 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Created on Mon Feb 24 12:03:11 2025
-
-@author: MIPO10053340
-"""
-#JWT
-from dotenv import load_dotenv
-load_dotenv()
-
-import os
-import numpy as np
-import pandas as pd
-from scipy.stats import entropy
-
-# API Clients
-from mistralai import Mistral
-from openai import OpenAI
-
-# from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
-
-
-# ⚙️ Configurations API (remplace par tes clés API)
-MISTRAL_API_KEY = os.getenv('MISTRAL_API_KEY_static')
-OPENAI_API_KEY = os.getenv('OPENAI_API_KEY_static')
-# ANTHROPIC_API_KEY = os.getenv('ANTHROPIC_API_KEY_static')
-# LLAMA_API_KEY = os.getenv('LLAMA_API_KEY_static')
-# HUGGINGFACE_TOKEN =os.getenv('HUGGINGFACE_TOKEN_static')
-
-# 📌 Choix des modèles à utiliser
-USE_MODELS = {
-    "mistral": True,
-    "gpt-4": False,
-    "llama": False,  # Active si tu veux l'utiliser
-    "qwen": False, 
-    "deepseek": False
-}
-
-
-# 📊 Fonction pour calculer l'entropie des réponses
-def calculate_entropy(text):
-    tokens = text.split()
-    probas = np.array([tokens.count(word) / len(tokens) for word in set(tokens)])
-    return entropy(probas)
-
-# 🚀 Fonction pour interroger les modèles
-def get_model_responses(question):
-    responses = {}
-
-    # # 🔹 MISTRAL
-    if USE_MODELS["mistral"]:
-        # Initialisation du client Mistral
-        client = Mistral(api_key=MISTRAL_API_KEY)
-        
-        # Créer une complétion de chat
-        response = client.chat.complete(
-            model="mistral-medium",
-            messages=[
-                {"role": "user", "content": question}
-            ]
-        )
-        # Extraire et afficher la réponse
-        text_response = response.choices[0].message.content    
-        responses["mistral-medium"] = {"response": text_response, "entropy": calculate_entropy(text_response)}
-
-    # 🔹 GPT-4 (OpenAI)
-    if USE_MODELS["gpt-4"]:
-        # Initialisation du client OpenAI
-        client = OpenAI(api_key=OPENAI_API_KEY)
-        
-        # Créer une complétion de chat
-        response = client.chat.completions.create(
-            model="gpt-4-turbo",
-            messages=[
-                {"role": "user", "content": question}
-            ]
-        )
-        # Extraire et afficher la réponse
-        text_response = response.choices[0].message.content    
-        responses["gpt-4-turbo"] = {"response": text_response, "entropy": calculate_entropy(text_response)}
-
-    # # 🔹 LLAMA (Hugging Face)
-    # if USE_MODELS["llama"]:
-    #     model_id = "meta-llama/Llama-2-7b-chat-hf"
-    #     tokenizer = AutoTokenizer.from_pretrained(model_id, use_auth_token=HUGGINGFACE_TOKEN)
-    #     model = AutoModelForCausalLM.from_pretrained(model_id, use_auth_token=HUGGINGFACE_TOKEN)
-    #     pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
-    
-    #     text_response = pipe(question, max_length=300)[0]["generated_text"]
-    #     responses["llama"] = {"response": text_response, "entropy": calculate_entropy(text_response)}
-    # # 🔹 QWEN (Hugging Face)
-    # if USE_MODELS["qwen"]:
-    #     model_id = "Qwen/Qwen-7B-Chat"
-    #     tokenizer = AutoTokenizer.from_pretrained(model_id, token=HUGGINGFACE_TOKEN, trust_remote_code=True)
-    #     model = AutoModelForCausalLM.from_pretrained(model_id, token=HUGGINGFACE_TOKEN, trust_remote_code=True)
-    #     pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
-    #     text_response = pipe(question, max_length=300)[0]["generated_text"]
-    #     responses["qwen"] = {"response": text_response, "entropy": calculate_entropy(text_response)}
-
-    # # 🔹 DEEPSEEK (Hugging Face)
-    # if USE_MODELS["deepseek"]:
-    #     model_id = "deepseek-ai/deepseek-7b-chat"
-    #     tokenizer = AutoTokenizer.from_pretrained(model_id)
-    #     model = AutoModelForCausalLM.from_pretrained(model_id)
-    #     pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
-    #     text_response = pipe(question, max_length=300)[0]["generated_text"]
-    #     responses["deepseek"] = {"response": text_response, "entropy": calculate_entropy(text_response)}
-
-    return responses
-
-# 📌 Question de test
-question = "Quels sont les besoins en protéines des poulets de chair en phase de croissance ?"
-
-# 🔥 Exécuter le test
-results = get_model_responses(question)
-
-# 📊 Afficher les résultats
-df = pd.DataFrame.from_dict(results, orient="index")
-print(df)
-
-# 💾 Sauvegarde en .txt
-with open("model_responses.txt", "w", encoding="utf-8") as f:
-    for model, data in results.items():
-        f.write(f"🔹 Modèle : {model.upper()}\n")
-        f.write(f"Réponse :\n{data['response']}\n")
-        f.write(f"📊 Entropie : {data['entropy']:.4f}\n")
-        f.write("=" * 50 + "\n\n")
-
-print("\n✅ Réponses enregistrées dans 'model_responses.txt'")
-