Hosting your Models and Datasets on Hugging Face Spaces using Streamlit

Showcase your Datasets and Models using Streamlit on Hugging Face Spaces

Streamlit allows you to visualize datasets and build demos of Machine Learning models in a neat way. In this blog post we will walk you through hosting models and datasets and serving your Streamlit applications in Hugging Face Spaces.

Building demos for your models

You can load any Hugging Face model and build cool UIs using Streamlit. In this particular example we will recreate "Write with Transformer" together. It's an application that lets you write anything using transformers like GPT-2 and XLNet.

We will not dive deep into how the inference works. You only need to know that you need to specify some hyperparameter values for this particular application. Streamlit provides many components for you to easily implement custom applications. We will use some of them to receive necessary hyperparameters inside the inference code.

• The .text_area component creates a nice area to input sentences to be completed.
• The Streamlit .sidebar method enables you to accept variables in a sidebar.
• The slider is used to take continuous values. Don't forget to give slider a step, otherwise it will treat the values as integers.
• You can let the end-user input integer vaues with number_input .

The inference code returns the generated output, you can print the output using simple st.write. st.write(generated_sequences[-1])

Here's what our replicated version looks like.

You can checkout the full code here.

Showcase your Datasets and Data Visualizations

Streamlit provides many components to help you visualize datasets. It works seamlessly with 🤗 Datasets, pandas, and visualization libraries such as matplotlib, seaborn and bokeh.

Let's start by loading a dataset. A new feature in Datasets, called streaming, allows you to work immediately with very large datasets, eliminating the need to download all of the examples and load them into memory.

st.write("Most appearing words including stopwords")
st.bar_chart(words[0:50])