Chatbots are widely studied in natural language processing (NLP) research and are a common use case of NLP in industry. Because chatbots are designed to be used directly by customers and end users, it is important to validate that chatbots are behaving as expected when confronted with a wide variety of input prompts.
Using `gradio`, you can easily build a demo of your chatbot model and share that with a testing team, or test it yourself using an intuitive chatbot GUI.
This tutorial will show how to take a pretrained chatbot model and deploy it with a Gradio interface in 4 steps. The live chatbot interface that we create will look something like this (try it!):
Chatbots are *stateful*, meaning that the model's prediction can change depending on how the user has previously interacted with the model. So, in this tutorial, we will also cover how to use **state** with Gradio demos.
Make sure you have the `gradio` Python package already [installed](/quickstart). To use a pretrained chatbot model, also install `transformers` and `torch`.
First, you will need to have a chatbot model that you have either trained yourself or you will need to download a pretrained model. In this tutorial, we will use a pretrained chatbot model, `DialoGPT`, and its tokenizer from the [Hugging Face Hub](https://huggingface.co/microsoft/DialoGPT-medium), but you can replace this with your own model.
Here is the code to load `DialoGPT` from Hugging Face `transformers`.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
Let's break this down. The function takes two parameters:
*`input`: which is what the user enters (through the Gradio GUI) in a particular step of the conversation.
*`history`: which represents the **state**, consisting of the list of user and bot responses. To create a stateful Gradio demo, we *must* pass in a parameter to represent the state, and we set the default value of this parameter to be the initial value of the state (in this case, the empty list since this is what we would like the chat history to be at the start).
Then, the function tokenizes the input and concatenates it with the tokens corresponding to the previous user and bot responses. Then, this is fed into the pretrained model to get a prediction. Finally, we do some cleaning up so that we can return two values from our function:
*`response`: which is a list of tuples of strings corresponding to all of the user and bot responses. This will be rendered as the output in the Gradio demo.
*`history` variable, which is the token representation of all of the user and bot responses. In stateful Gradio demos, we *must* return the updated state at the end of the function.
Now that we have our predictive function set up, we can create a Gradio Interface around it.
In this case, our function takes in two values, a text input and a state input. The corresponding input components in `gradio` are `"text"` and `"state"`.
The function also returns two values. We will display the list of responses using the dedicated `"chatbot"` component and use the `"state"` output component type for the second return value.
* The final [chatbot demo](https://huggingface.co/spaces/abidlabs/chatbot-stylized) and [complete code](https://huggingface.co/spaces/abidlabs/chatbot-stylized/tree/main) (on Hugging Face Spaces)
