update

app.py

@@ -1,18 +1,17 @@
-from transformers import AutoModel, AutoTokenizer
+from transformers import  AutoTokenizer,AutoModel
 import torch
-
-tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True)
-#model = AutoModel.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True).half().cuda()
-model = AutoModel.from_pretrained("THUDM/chatglm-6b-int4",trust_remote_code=True).float()
-
+chat_tkn = AutoTokenizer.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True)
+mdl = AutoModel.from_pretrained("THUDM/chatglm-6b-int4",trust_remote_code=True).float()
+#chat_tkn = AutoTokenizer.from_pretrained("facebook/blenderbot-400M-distill")
+#mdl = BlenderbotForConditionalGeneration.from_pretrained("facebook/blenderbot-400M-distill")
 def converse(user_input, chat_history=[]):
-    user_input_ids = tokenizer(user_input + tokenizer.eos_token, return_tensors='pt').input_ids
+    user_input_ids = chat_tkn(user_input + chat_tkn.eos_token, return_tensors='pt').input_ids
     # keep history in the tensor
     bot_input_ids = torch.cat([torch.LongTensor(chat_history), user_input_ids], dim=-1)
     # get response
-    chat_history = model.generate(bot_input_ids, max_length=1000, pad_token_id=tokenizer.eos_token_id).tolist()
+    chat_history = mdl.generate(bot_input_ids, max_length=1000, pad_token_id=chat_tkn.eos_token_id).tolist()
     print (chat_history)
-    response = tokenizer.decode(chat_history[0]).split("<|endoftext|>")
+    response = chat_tkn.decode(chat_history[0]).split("<|endoftext|>")
     print("starting to print response")
     print(response)
     # html for display