The KDnuggets Gradio Crash Course

Image by Editor

# Introducing Gradio

 
Gradio is a Python framework that changes how machine learning practitioners create interactive web interfaces for their models. With just a few lines of code, you can build polished applications that accept various inputs (text, images, audio) and display outputs in an intuitive way. Whether you’re a researcher, data scientist, or developer, Gradio makes model deployment accessible to everyone.

Some of the benefits of Gradio include:

• It allows you to go from model to demo in minutes
• You don’t need frontend skills, just pure Python implementation
• It has support for text, images, audio, and more
• You can easily share and deploy locally, and can also host publicly for free

# Installing Gradio and Basic Setup

 
To get started with Gradio, you need to install the package using the pip command.

Now that you have Gradio installed, let’s create your first Gradio application. First, create a file and name it gradio_app.py then add this code:

import gradio as gr
def greet(name):
    return f"Hello {name}!"

demo = gr.Interface(
    fn=greet,
    inputs="text",
    outputs="text",
    title="Greeting App"
)

demo.launch()

Run this with python gradio_app.py, and you’ll have a running web application at http://127.0.0.1:7860/. The interface provides a text input, a submit button, and a text output — all automatically generated from your simple specification.

Image by Author

// Understanding the Gradio Interface

The gr.Interface class is Gradio’s high-level application programming interface (API) that abstracts away complexity. It requires three essential components:

• Function (fn): Your Python function that processes inputs
• Inputs: Specification of input type(s)
• Outputs: Specification of output type(s)

// Exploring Input and Output Components

While you can use simple strings like "text", "image", or "audio" to specify components, Gradio offers more control through explicit component classes.

import gradio as gr

demo = gr.Interface(
    fn=lambda x: x,
    inputs=gr.Textbox(lines=2, placeholder="Enter text here..."),
    outputs=gr.Textbox(label="Output")
)

Common components include:

• gr.Textbox(): Multi-line text input
• gr.Image(): Image upload/preview
• gr.Audio(): Audio file handling
• gr.Checkbox(): Boolean input
• gr.Slider(): Numerical range input
• gr.Radio(): Multiple choice selection
• gr.Dropdown(): Select from options

// Handling Multiple Inputs and Outputs

Real-world applications often require multiple inputs or produce multiple outputs. Gradio handles this elegantly with lists.

import gradio as gr

def process_form(name, is_morning, temperature):
    greeting = "Good morning" if is_morning else "Hello"
    message = f"{greeting}, {name}! Temperature: {temperature}°C"
    return message, temperature * 1.8 + 32  # Convert to Fahrenheit

demo = gr.Interface(
    fn=process_form,
    inputs=[
        gr.Textbox(label="Name"),
        gr.Checkbox(label="Is it morning?"),
        gr.Slider(0, 100, label="Temperature (°C)")
    ],
    outputs=[
        gr.Textbox(label="Greeting"),
        gr.Number(label="Temperature (°F)")
    ]
)

demo.launch()

Output:

Image by Author

When using multiple inputs, your function must accept the same number of parameters. Similarly, multiple outputs require your function to return multiple values.

// Processing Images

Gradio makes image processing models incredibly easy to demo:

import gradio as gr
import numpy as np

def apply_sepia(image):
    # Image comes as numpy array with shape (height, width, channels)
    sepia_filter = np.array([[0.393, 0.769, 0.189],
                             [0.349, 0.686, 0.168],
                             [0.272, 0.534, 0.131]])
    sepia_image = image.dot(sepia_filter.T)
    sepia_image = np.clip(sepia_image, 0, 255).astype(np.uint8)
    return sepia_image

demo = gr.Interface(
    fn=apply_sepia,
    inputs=gr.Image(label="Input Image"),
    outputs=gr.Image(label="Sepia Filtered"),
    title="Sepia Filter App"
)

demo.launch()

Output:

Image by Author

The gr.Image component automatically handles file uploads, previews, and converts images to NumPy arrays for processing.

// Handling Audio Processing

Audio applications are just as straightforward:

import gradio as gr

def transcribe_audio(audio):
    return "Transcribed text would appear here"

demo = gr.Interface(
    fn=transcribe_audio,
    inputs=gr.Audio(label="Upload Audio", type="filepath"),
    outputs=gr.Textbox(label="Transcription"),
    title="Speech-to-Text Demo"
)
demo.launch()

In a real application, you would call a speech recognition model inside the transcribe_audio(audio) function. For demonstration, we’ll return a placeholder.

Output:

Image by Author

# Creating Advanced Layouts with Gradio Blocks

 
While gr.Interface is perfect for simple applications, gr.Blocks offers complete control over layout and data flow. Think of Blocks as the low-level API that lets you build complex, multi-step applications.

// Implementing a Basic Blocks Example

import gradio as gr

def greet(name):
    return f"Hello {name}!"

with gr.Blocks() as demo:
    name_input = gr.Textbox(label="Your Name")
    greet_button = gr.Button("Greet")
    output = gr.Textbox(label="Greeting")
    
    greet_button.click(
        fn=greet,
        inputs=name_input,
        outputs=output
    )

demo.launch()

Output:

Image by Author

// Building Complex Layouts with Rows and Columns

Here’s a more sophisticated example integrating with Transformers. Ensure that the Transformers package is installed on your computer.

pip install transformers

import gradio as gr
from transformers import pipeline

# Load a translation model
translator = pipeline("translation_en_to_de", model="t5-small")

def translate_text(text):
    result = translator(text, max_length=40)[0]
    return result['translation_text']

with gr.Blocks(title="English to German Translator") as demo:
    gr.Markdown("# 🌍 English to German Translator")
    
    with gr.Row():
        with gr.Column():
            english_input = gr.Textbox(
                label="English Text",
                placeholder="Enter text to translate...",
                lines=4
            )
            translate_btn = gr.Button("Translate", variant="primary")
        
        with gr.Column():
            german_output = gr.Textbox(
                label="German Translation",
                lines=4
            )

    # Add example prompts
    gr.Examples(
        examples=[
            ["Hello, how are you?"],
            ["The weather is beautiful today"],
            ["Machine learning is fascinating"]
        ],
        inputs=english_input
    )
    
    translate_btn.click(
        fn=translate_text,
        inputs=english_input,
        outputs=german_output
    )

demo.launch()

Output:

Image by Author

# Managing State in Gradio Applications

 
State management is important for interactive applications. Gradio provides two approaches: global state and session state.

// Managing Session State (User-Specific)

For user-specific state, use Gradio’s built-in state management. The following example demonstrates a simple chatbot logic using state to maintain conversation history.

import gradio as gr

with gr.Blocks() as demo:
    chatbot = gr.Chatbot(label="Conversation")
    msg = gr.Textbox(label="Your Message")
    clear = gr.Button("Clear")
    
    state = gr.State([])
    
    def user_message(message, history):
        # Update history with user message and placeholder for bot
        return "", history + [[message, None]]
    
    def bot_response(history):
        # Simple echo bot logic
        response = f"I received: {history[-1][0]}"
        history[-1][1] = response
        return history
    
    msg.submit(
        user_message,
        [msg, state],
        [msg, state]
    ).then(
        bot_response,
        state,
        chatbot
    )
    
    clear.click(lambda: (None, []), None, [chatbot, state])

demo.launch()

# Deploying and Sharing Your Applications

 
For quick sharing, Gradio can create a public URL:

This generates a temporary, publicly accessible link perfect for demos and quick sharing with colleagues. It is typically valid for 72 hours.

For free, permanent hosting:

• Create a Hugging Face account
• Create a new Space with Gradio as the software development kit (SDK)
• Upload your application files: app.py (your main application file) and requirements.txt (Python dependencies). An example of what should be in the requirements.txt file:

git add .
git commit -m "Initial commit"
git push

Your application will be available at https://huggingface.co/spaces/your-username/your-space-name.

Gradio applications can be deployed on any platform that supports Python web applications:

• Use demo.launch(server_name="0.0.0.0", server_port=7860)
• Package your application with all dependencies inside a Docker container
• Deploy on AWS, Google Cloud, Azure, and other platforms

# Building an Image Classification Dashboard

 
Putting everything we have learned together, let’s build a project. This project is a simple image classification dashboard built with PyTorch and Gradio. It enables users to upload an image through a web interface and receive the top five predicted classes generated by a pre-trained deep learning model.

We will use ResNet-50, a well-known convolutional neural network trained on the ImageNet dataset. Because the model is pre-trained, the project does not require any custom training or labeled data. It is intended for demonstration, experimentation, and educational purposes rather than production use.

We will use Gradio to provide a lightweight user interface so users can interact with the model directly from a browser.

import gradio as gr
import torch
from torchvision import models, transforms
from PIL import Image

# Load pre-trained model
model = models.resnet50(pretrained=True)
model.eval()

# Preprocessing
preprocess = transforms.Compose([
    transforms.Resize(256),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize(
        mean=[0.485, 0.456, 0.406],
        std=[0.229, 0.224, 0.225]
    )
])

def classify_image(image):
    image = Image.fromarray(image)
    input_tensor = preprocess(image)
    input_batch = input_tensor.unsqueeze(0)
    
    with torch.no_grad():
        output = model(input_batch)

    # Get top 5 predictions
    probabilities = torch.nn.functional.softmax(output[0], dim=0)
    top5_prob, top5_catid = torch.topk(probabilities, 5)
    
    results = []
    for i in range(top5_prob.size(0)):
        results.append(f"Category {top5_catid[i].item()}: {top5_prob[i].item()*100:.2f}%")
    
    return "\n".join(results)

demo = gr.Interface(
    fn=classify_image,
    inputs=gr.Image(label="Upload Image"),
    outputs=gr.Textbox(label="Top 5 Predictions"),
    title="Image Classifier"
)

demo.launch()

# Wrapping Up

 
Gradio makes machine learning deployment easy by eliminating the traditional barriers between model development and user interaction. With this crash course, you’ve learned the fundamentals of creating Gradio interfaces, component-based design for diverse input/output types, advanced layouts using Gradio Blocks, state management for interactive applications, and deployment strategies for sharing your work.

The true power of Gradio lies in its simplicity and flexibility. It doesn’t matter if you’re building a quick prototype for internal testing or a polished application for public use; Gradio provides the tools you need to bring your machine learning models to life.
 
 

Shittu Olumide is a software engineer and technical writer passionate about leveraging cutting-edge technologies to craft compelling narratives, with a keen eye for detail and a knack for simplifying complex concepts. You can also find Shittu on Twitter.