Upload app.py

app.py

@@ -1,31 +1,25 @@
+import openai
+from keys import mykey
 import gradio as gr
-from transformers import AutoTokenizer, AutoModelForCausalLM
-import torch
 
-# Load your model and tokenizer
-tokenizer = AutoTokenizer.from_pretrained("gpt2")
-model = AutoModelForCausalLM.from_pretrained("gpt2")
-model.eval()
+openai.api_key = mykey["mediQ"]
 
-# Define chat function
-def mediq_chat(message, history=[]):
-    # You can optionally join past conversation to provide more context
-    input_ids = tokenizer.encode(message, return_tensors="pt")
-    with torch.no_grad():
-        output = model.generate(
-            input_ids,
-            max_new_tokens=200,
-            do_sample=True,
-            temperature=0.7,
-            top_p=0.9
-        )
-    response = tokenizer.decode(output[0], skip_special_tokens=True)
-    return response
+def gpt_response(message, history=[]):
+    prompt = "\n".join(history + [message])
+    response = openai.ChatCompletion.create(
+        model="gpt-4",
+        messages=[
+            {"role": "system", "content": "You are a careful and experienced clinical reasoning expert."},
+            {"role": "user", "content": prompt}
+        ],
+        temperature=0.3,
+        max_tokens=300
+    )
+    reply = response["choices"][0]["message"]["content"]
+    return reply
 
-# Create the Gradio chat interface
 gr.ChatInterface(
-    fn=mediq_chat,
-    title="🧠 MediQ - Clinical Reasoning Chatbot",
-    description="Ask clinical questions. The model will respond based on its training.",
-    theme="default"
+    fn=gpt_response,
+    title="🧠 MediQ GPT-4 Clinical Reasoning",
+    description="This private app uses GPT-4 to simulate adaptive question-asking in clinical reasoning."
 ).launch()