Create chat.py

chat.py

@@ -0,0 +1,21 @@
+# chat.py
+import torch
+from model_loader import load_model
+from config import MAX_TOKENS, TEMPERATURE
+
+tokenizer, model = load_model()
+
+def generate_response(prompt, max_length=MAX_TOKENS, temperature=TEMPERATURE):
+    """
+    Generate a response from the model
+    """
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    outputs = model.generate(
+        **inputs,
+        max_length=max_length,
+        do_sample=True,
+        temperature=temperature,
+        pad_token_id=tokenizer.eos_token_id
+    )
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return response