Demo Application
Learn how the Hugging Face Inference API can be used in an example application.
We'll cover the following
In the previous lessons, we gained hands-on experience with the Hugging Face Inference API endpoints. Let's integrate those endpoints into a functional application.
Workflow
The workflow of the application is as follows:
When we run the application, the landing page is rendered and displays the page where we can do sentiment analysis.
We can move between different components using the side navigation bar. Each page uses a new endpoint and performs the mentioned task.
We can see the logo in the top-left corner and a sun/moon icon in the top-right corner. Clicking on the icon will change the theme to light/dark mode.
The widget below contains the code for our application. Click the "Run" button, then click the link at the end of the widget to see the React application in action.
.home{ display: flex; flex: 1; justify-content: center; background-color: var(--bg-clr); position: fixed; width: 90%; height: 100%; padding:0; padding-top: 5rem; padding-bottom: 5rem; margin:0; margin-left: 10%; padding-left: 4%; top:0; /* left:15rem; */ overflow-y: scroll; } .title{ font-size: 4rem; font-weight: 800; color: var(--title-clr); margin-top: 5rem; text-align: center; } .input-box{ width: 90%; height: 70%; margin: auto; margin-top: 2rem; border-radius: 1rem; border: 1px solid var(--input-box-border-clr); background-color: var(--input-box-bg-clr); } .ib{ width: 70%; height: 70%; margin: auto; border-radius: 1rem; border: 1px solid var(--input-box-border-clr); display: flex; flex-direction: column; justify-content: center; } .input-box.summary{ height: 40%; } .input-box.translation{ width: 45%; } .input-box.context{ margin: 0rem; margin-right: 5rem; width: 70%; } .input-box.question{ width: 100%; height: 45%; margin: 0rem; margin-bottom: 5rem; } .input-box.answer{ width: 100%; height: 60%; margin: 0rem; } .input-box.similarity{ width: 97%; height: 49%; } .input-box.fillmask{ width: 97%; } .input-box.classify{ width: 97%; height: 7rem; } .input-box.option{ width: 25%; height: 5rem; } .input-area{ width: 100%; height: 88%; border: none; background-color: transparent; font-size: 2.5rem; padding: 1rem; color: var(--title-clr); } .input-area.question{ height: 78%; } .input-area.similarity{ height: 100%; } .input-area.fill-mask{ width: 80%; font-size: 3rem; margin: auto; } .input-area::placeholder{ color: var(--placeholder-clr); } .input-area:focus{ border: none; outline: none; } .fill-mask{ display: flex; justify-content: center; width: 100%; height: 100%; align-items: center; color: var(--title-clr); font-size: 4rem; } .button{ width: 15rem; height: 4.7rem; font-size: 2rem; font-weight: bold; color: #fff; background-color: var(--nav-clr); padding: 1rem; border: none; border-radius: 0.5rem; margin: 0rem 0.5rem; cursor: pointer; position: absolute; bottom: 0; } .button.t{ border: none; color: var(--title-clr); background-color: transparent; width: 7rem; font-size: 2.2rem; font-weight: bold; } .button:hover{ /* background-color: var(--title-clr); color: var(--bg-clr); */ /* box-shadow: 0 4px 8px 0 rgba(223, 223, 177, 0.2), 0 6px 20px 0 rgba(0, 0, 0, 0.19); */ } .button.copy{ width: 18rem; border: 0.2rem solid var(--nav-clr); color: var(--nav-clr); background-color: transparent; /* margin-bottom: 0.5rem; */ } .btn1{ width: 15rem; height: 4.7rem; font-size: 2rem; font-weight: bold; color: #fff; background-color: var(--nav-clr); padding: 1rem; border: none; border-radius: 0.5rem; margin: auto 0rem; cursor: pointer; margin-top: 1rem; text-align: center; } .btn1:hover{ /* background-color: var(--title-clr); color: var(--bg-clr); box-shadow: 0 4px 8px 0 rgba(223, 223, 177, 0.2), 0 6px 20px 0 rgba(0, 0, 0, 0.19); */ } .btn1.copy{ width: 18rem; margin: auto; border: 0.1rem solid var(--nav-clr); color: var(--nav-clr); background-color: transparent; } .btn1.s{ border: 0.2rem solid var(--nav-clr); color: var(--nav-clr); background-color: transparent; } .btn1.t{ border: none; color: var(--title-clr); background-color: transparent; width: 7rem; font-size: 2.2rem; font-weight: bold; } .btn1.copy.img{ width: 15rem; background-color: var(--nav-clr); color: #fff; margin: auto; } .btn1.upload{ margin: 0rem auto; } .emo{ font-size: 8rem; opacity: 0.4; margin: 3rem; } .score{ width: 15rem; height: 4.7rem; font-size: 2rem; font-weight: bold; color: #fff; background-color: var(--nav-clr); padding: 1rem; border: none; border-radius: 1rem; text-align: center; outline: none; margin: 0rem 0.5rem; position: absolute; bottom: 0; /* margin-bottom: 0.5rem; */ } .score-box{ width: 9rem; height: 9rem; border-radius: 1rem; background-color: var(--nav-clr); color: #fff; padding: 2rem; font-size: 4rem; text-align: center; opacity: 0.4; } .score-box-text{ width: 25rem; height: 9rem; color: var(--nav-clr); background-color: var(--bg-clr); border-radius: 0rem; font-size: 4rem; text-align: center; padding: 1.7rem 1rem; opacity: 0.4; } .img-display{ display: block; width: 90%; height: 90%; border-radius: 2rem; margin: auto; top: 0; left: 0; right: 0; bottom: 0; padding: 0; background: rgba(0, 0, 0, 0.3); box-shadow: 0 4px 8px 0 rgba(223, 223, 177, 0.2), 0 6px 20px 0 rgba(0, 0, 0, 0.19); } .res-img-display{ height: 100%; border-radius: 2rem; margin: auto; top: 0; left: 0; right: 0; bottom: 0; padding: 0; background: rgba(0, 0, 0, 0.3); box-shadow: 0 4px 8px 0 rgba(223, 223, 177, 0.2), 0 6px 20px 0 rgba(0, 0, 0, 0.19); } .dropdown { position: relative; display: inline-block; } .dropdown-content { display: none; position: absolute; background-color: var(--nav-clr); border-radius: 1rem; bottom: 5rem; min-width: 160px; /* box-shadow: 0px 8px 16px 0px rgba(0,0,0,0.2); */ z-index: 1; } .dropdown-content span { color: #fff; font-size: 2rem; padding: 12px 16px; display: block; cursor: pointer; text-align: center; width: 16rem; } .dropdown-content a:hover {background-color: #ddd;} .dropdown:hover .dropdown-content {display: block;} input::-webkit-outer-spin-button, input::-webkit-inner-spin-button { -webkit-appearance: none; margin: 0; } input[type=number]{ -moz-appearance: textfield; }
Code explanation
Let’s dive into the code and see how we’ve integrated different Hugging Face Inference API endpoints into our React application. Our application contains various components, each of which performs a specific function, as described below:
The
Navbar
component renders the navigation bar, which contains the logo and an icon that allows the user to switch between light and dark modes.The
Error
component is responsible for displaying errors.The
Loading
component renders the customized loader.The
SentimentAnalysis
,Summarization
,Translation
,QuestionAnswering
,TextGeneration
,SentenceSimilarity
, andFillMask
components are responsible for rendering pages where Hugging Face NLP-related endpoints are called.Each page provides fields to the user depending on the respective endpoint parameter's requirement and displays the result.
The
Summarization
andTextGeneration
components have a cog on the left of the summarize/generate button. Clicking it allows us to change optional parameters to get optimal results.The
Translation
component allows us to select between six different languages for text translation.
The
ImageClassification
,ObjectDetection
, andSegmentation
components display pages where Hugging Face CV-related endpoints are used.Each page allows the user to upload the image using the "Upload File" button or an image URL and then displays the response returned by the API call.