should be done better but it works

main.py

@@ -17,19 +17,19 @@ from transformers.models.auto.modeling_auto import AutoModelForCausalLM
 tokenizer = AutoTokenizer.from_pretrained("microsoft/DialoGPT-large")
 model = AutoModelForCausalLM.from_pretrained("../southpark/output-medium")
 
-def cartman_respond(input_text):
+def cartman_respond(packet):
-    input_ids = tokenizer(input_text + tokenizer.eos_token, return_tensors="pt").input_ids
+    input_ids = tokenizer(str(packet.get('message')) + tokenizer.eos_token, return_tensors="pt").input_ids
     outputs = model.generate(
             input_ids,
             pad_token_id=tokenizer.eos_token_id,
-            max_new_tokens = 200,
+            max_new_tokens = int(packet.get('max_new_tokens')),
-            num_beams = 8,
+            num_beams = int(packet.get('num_beams')),
-            num_beam_groups = 4,
+            num_beam_groups = int(packet.get('num_beam_groups')),
-            no_repeat_ngram_size=3,
+            no_repeat_ngram_size = int(packet.get('no_repeat_ngram_size')),
-            length_penalty = 1.4,
+            length_penalty = float(packet.get('length_penalty')),
-            diversity_penalty = 0,
+            diversity_penalty = float(packet.get('diversity_penalty')),
-            repetition_penalty = 2.1,
+            repetition_penalty = float(packet.get('repetition_penalty')),
-            early_stopping = True,
+            early_stopping = bool(packet.get('early_stopping')),
 
           # do_sample = True,
           # top_k = 100,
@@ -41,9 +41,10 @@ def cartman_respond(input_text):
 @api.post('/chat/')
 async def getInformation(data : Request):
     packet = await data.json()
-    message = packet.get('Message')
+    print(packet)
+    message = str(packet.get('message'))
     print(message)
-    response = cartman_respond(message)
+    response = cartman_respond(packet)
     print(response)
 
     return {