Update app.py

app.py

@@ -1,14 +1,125 @@
-import torch
-from peft import PeftModel, PeftConfig
-from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig, pipelin
-bnb_config = BitsAndBytesConfig(load_in_4bit=True,
-    bnb_4bit_quant_type="nf4",
-    bnb_4bit_compute_dtype=torch.bfloat16,
-    bnb_4bit_use_double_quant=False)
-     
-model_id = "meta-llama/Meta-Llama-3-8B"
-tokenizer = AutoTokenizer.from_pretrained(model_id)
-model = AutoModelForCausalLM.from_pretrained(model_id, quantization_config = bnb_config,device_map={"":0})
-
-
-     
+import gradio as gr
+import cohere
+import os
+import re
+import uuid
+import secrets
+
+
+
+Mistralai = "19NAIlRYNoVaNYoNoxkdq1mCmbdAgxaa"
+co = cohere.Client(Mistralai, client_name="huggingface-rp")
+
+
+def trigger_example(example):
+    chat, updated_history = generate_response(example)
+    return chat, updated_history
+        
+def generate_response(user_message, cid, token, history=None):
+
+    if not token:
+        raise gr.Error("Error loading.")
+        
+    if history is None:
+        history = []
+    if cid == "" or None:    
+        cid = str(uuid.uuid4())
+
+    print(f"cid: {cid} prompt:{user_message}")
+    
+    history.append(user_message)
+    
+    stream = co.chat_stream(message=user_message, conversation_id=cid, model='Codestral-22B-v0.1', connectors=[], temperature=0.3)
+    
+    output = ""
+    
+    for idx, response in enumerate(stream):
+        if response.event_type == "text-generation":
+            output += response.text
+        if idx == 0:
+            history.append(" " + output)
+        else:
+            history[-1] = output
+        chat = [
+            (history[i].strip(), history[i + 1].strip())
+            for i in range(0, len(history) - 1, 2)
+        ] 
+        yield chat, history, cid
+        
+    return chat, history, cid
+    
+
+def clear_chat():
+    return [], [], str(uuid.uuid4())
+
+
+examples = [
+    "Pouvez-vous expliquer les règles relatives aux heures supplémentaires selon le Code du travail ?",
+    "Quels sont les droits et obligations concernant les congés payés selon le Code du travail ?",
+    "Quelles sont les conditions requises pour un licenciement pour motif économique en vertu du Code du travail ?",
+    "Quelles sont les mesures prévues par le Code du travail en cas de harcèlement au travail ?",
+   
+]
+
+custom_css = """
+#logo-img {
+    border: none !important;
+}
+#chat-message {
+    font-size: 14px;
+    min-height: 300px;
+}
+"""
+
+with gr.Blocks(analytics_enabled=False, css=custom_css) as demo:
+    cid = gr.State("")
+    token = gr.State(value=None)
+    
+    with gr.Row():
+        with gr.Column(scale=1):
+            gr.Image("logohf1.jpg", elem_id="logo-img", show_label=False, show_share_button=False, show_download_button=False)
+        with gr.Column(scale=3):
+            gr.Markdown("""##### Bienvenue sur votre Assistant Code du Travail Français ! """)
+            
+    with gr.Column():
+        with gr.Row():
+            chatbot = gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=True)
+        
+        with gr.Row():
+            user_message = gr.Textbox(lines=1, placeholder="Question ...", label="Input", show_label=False)
+
+      
+        with gr.Row():
+            submit_button = gr.Button("Envoyer")
+            clear_button = gr.Button("Supprimer la discussion")
+
+                        
+        history = gr.State([])
+        
+        user_message.submit(fn=generate_response, inputs=[user_message, cid, token, history], outputs=[chatbot, history, cid], concurrency_limit=32)
+        submit_button.click(fn=generate_response, inputs=[user_message, cid, token, history], outputs=[chatbot, history, cid], concurrency_limit=32)
+        
+        clear_button.click(fn=clear_chat, inputs=None, outputs=[chatbot, history, cid], concurrency_limit=32)
+
+        user_message.submit(lambda x: gr.update(value=""), None, [user_message], queue=False)
+        submit_button.click(lambda x: gr.update(value=""), None, [user_message], queue=False)
+        clear_button.click(lambda x: gr.update(value=""), None, [user_message], queue=False)
+        
+        with gr.Row():
+            gr.Examples(
+                examples=examples,
+                inputs=user_message,
+                cache_examples=False,
+                fn=trigger_example,
+                outputs=[chatbot],
+                examples_per_page=100
+            )
+
+    demo.load(lambda: secrets.token_hex(16), None, token)
+
+if __name__ == "__main__":
+    # demo.launch(debug=True)
+    try:
+        demo.queue(api_open=False, max_size=40).launch(show_api=False)
+    except Exception as e:
+        print(f"Error: {e}")