Update app.py

app.py

@@ -1,68 +1,56 @@
-import gradio as gr
-from huggingface_hub import InferenceClient
 from huggingface_hub import login
+import torch
+import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-# # Authentification avec Hugging Face
-# login(token=os.getenv('API_KEY'))
-# print(os.getenv('API_KEY'))
-
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("mistralai/mathstral-7B-v0.1")
-
-
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
+# Authentification avec Hugging Face
+login(token=os.getenv('API_KEY'))
+print(os.getenv('API_KEY'))
 
-    response = ""
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+# Charger le modèle et le tokenizer
+model_name = "EleutherAI/mathstral-7b-v1"  # Remplacez par le nom exact du modèle Mathstral
+tokenizer = AutoTokenizer.from_pretrained(model_name)
 
-        response += token
-        yield response
+# Charger le modèle avec quantification dynamique
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype=torch.float16,
+    device_map="auto",
+    low_cpu_mem_usage=True,
+)
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+# Appliquer la quantification dynamique
+model = torch.quantization.quantize_dynamic(
+    model, {torch.nn.Linear}, dtype=torch.qint8
 )
 
+# Fonction pour générer du texte
+def generate_text(prompt, max_new_tokens=50):
+    inputs = tokenizer(prompt, return_tensors="pt")
+    with torch.no_grad():
+        outputs = model.generate(
+            **inputs, 
+            max_new_tokens=max_new_tokens, 
+            do_sample=True, 
+            top_p=0.95, 
+            temperature=0.7
+        )
+    return tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+# Interface Gradio
+def gradio_interface(prompt):
+    response = generate_text(prompt)
+    return response
+
+# Création de l'interface Gradio
+iface = gr.Interface(
+    fn=gradio_interface,
+    inputs=gr.Textbox(lines=2, placeholder="Entrez votre question mathématique ici..."),
+    outputs="text",
+    title="Assistant Mathématique (Mathstral-7B)",
+    description="Posez une question mathématique, et l'assistant vous répondra.",
+)
 
-if __name__ == "__main__":
-    demo.launch()
+# Lancer l'interface
+iface.launch()