Hugging Face's logo

Spaces:
Samarth991
/
GradCam_Explainable-AI
Sleeping

App Files Community
GradCam_Explainable-AI / README.md
Samarth991's picture
Samarth991
adding GradCAM for explainable AI
cf2d943
preview code
|
raw
history blame contribute delete
2.31 kB

A newer version of the Gradio SDK is available: 5.42.0

Upgrade
metadata
title: GradCAM CIFAR10
emoji: 😻
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 3.39.0
app_file: app.py
pinned: false
license: mit

CIFAR10 demo with GradCAM

How to Use the App

  1. The app has two tabs:

    • GradCAM: In this tab, you can look Visualize Class Activations Maps (helps to see what the model is actually looking at in the image) generated by the model’s layer for the predicted class
      • see existing GradCAM images (from stored misclassified images)
      • upload your own 32x32 pixel image or choose an example image provided to classify and visualize the class activation maps using GradCAM. You can adjust the number of top predicted classes, show/hide the GradCAM overlay, select target layer, and control the transparency of the overlay.
    • Misclassified Examples: In this tab, the app displays a gallery of misclassified images from CIFAR10 test dataset. You can control the number of examples shown

  2. GradCAM Tab:

    • View Existing Images:
      • Number of Images Select number of images to show, default is 1 and max is 10
      • Layers Select the target layers for GradCAM visualization
      • Opacity: Control the Opacity of the GradCAM overlay. The default value is 0.7.
    • New of Example Images
      • Input Image: Upload your own 32x32 pixel image or select one of the example images from the given list.
      • Top Classes: Choose the number of top predicted classes to display along with their respective confidence scores.
      • Enable GradCAM: Check this box to display the GradCAM overlay on the input image.
      • Network Layers: Select the target layers for GradCAM visualization.
      • Opacity: Control the Opacity of the GradCAM overlay. The default value is 0.7.

  3. Misclassified Examples Tab:

    • No. of Examples: Control the number of misclassified examples displayed in the gallery. The default value is 1, max is 10.

  4. After adjusting the settings, click the "Submit" button to see the results.

Training code

The main code using which training was performed can be viewed at below location:

https://github.com/peeyushsinghal/ERA/tree/main/S12

License

This project is licensed under the MIT License