update

app.py

@@ -1,18 +1,13 @@
+from transformers import GPT2Tokenizer, GPT2LMHeadModel
 from langchain import PromptTemplate, LLMChain
-from langchain.llms import GPT4All
 from langchain.memory import ConversationBufferMemory
 from langchain.chains import ConversationChain
 from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
-from huggingface_hub import hf_hub_download
-from transformers import AutoTokenizer, AutoModelForCausalLM
 import gradio as gr
 
-# model_path = hf_hub_download(repo_id="microsoft/DialoGPT-medium", filename="pytorch_model.bin")
 # Load the tokenizer and model
-tokenizer = AutoTokenizer.from_pretrained("microsoft/DialoGPT-medium")
-
-model = AutoModelForCausalLM.from_pretrained("microsoft/DialoGPT-medium")
-
+tokenizer = GPT2Tokenizer.from_pretrained("gpt2")
+model = GPT2LMHeadModel.from_pretrained("gpt2")
 
 template = """Question: {question}
 ------------------
@@ -24,8 +19,8 @@ prompt = PromptTemplate(template=template, input_variables=["question"])
 memory = ConversationBufferMemory()
 # Callbacks support token-wise streaming
 callbacks = [StreamingStdOutCallbackHandler()]
-# Verbose is required to pass to the callback manager
-llm = GPT4All(model=model, callbacks=callbacks, verbose=True)
+# Instantiate the LLMChain with the model and tokenizer
+llm = LLMChain(model=model, tokenizer=tokenizer, callbacks=callbacks, verbose=True)
 
 conversation = ConversationChain(llm=llm, memory=memory, callbacks=callbacks, prompt=prompt)
 

requirements.txt

@@ -1,5 +1,3 @@
 langchain==0.0.161
 gradio==2.3.0
-transformers
-torch
-pygpt4all
+transformers