Create app.py

app.py

@@ -0,0 +1,76 @@
+import transformers
+import torch
+import gradio as gr
+
+# Initialize the model and pipeline
+model_name_or_path = "m42-health/Llama3-Med42-8B"
+
+pipeline = transformers.pipeline(
+    "text-generation",
+    model=model_name_or_path,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
+
+# Define the system message for the chatbot personality
+system_message = {
+    "role": "system",
+    "content": (
+        "You are a helpful, respectful, and honest medical assistant. "
+        "You are a second version of Med42 developed by the AI team at M42, UAE. "
+        "Always answer as helpfully as possible, while being safe. "
+        "Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. "
+        "Please ensure that your responses are socially unbiased and positive in nature. "
+        "If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. "
+        "If you don’t know the answer to a question, please don’t share false information."
+    ),
+}
+
+# Define stop tokens
+stop_tokens = [
+    pipeline.tokenizer.eos_token_id,
+    pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>"),
+]
+
+# Initialize the conversation history
+conversation_history = [system_message]
+
+def chat_with_model(user_input):
+    # Append user message to conversation history
+    conversation_history.append({"role": "user", "content": user_input})
+
+    # Format the conversation for input to the model
+    prompt = pipeline.tokenizer.apply_chat_template(
+        conversation_history, tokenize=False, add_generation_prompt=False
+    )
+
+    # Generate response
+    outputs = pipeline(
+        prompt,
+        max_new_tokens=512,
+        eos_token_id=stop_tokens,
+        do_sample=True,
+        temperature=0.4,
+        top_k=150,
+        top_p=0.75,
+    )
+
+    # Extract the generated response (the part after the prompt)
+    generated_text = outputs[0]["generated_text"][len(prompt):]
+
+    # Append the assistant's response to the conversation history
+    conversation_history.append({"role": "assistant", "content": generated_text})
+
+    return generated_text.strip()
+
+# Create Gradio interface
+iface = gr.Interface(
+    fn=chat_with_model,
+    inputs="text",
+    outputs="text",
+    title="Med42 Medical Assistant",
+    description="Ask anything about medicine!",
+)
+
+# Launch the app
+iface.launch()