Spaces:
Runtime error
Runtime error
File size: 1,067 Bytes
0fd0025 7a4f7b8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 |
---
title: Pytorch Lighting
emoji: 🏢
colorFrom: purple
colorTo: indigo
sdk: gradio
sdk_version: 3.39.0
app_file: app.py
pinned: false
license: mit
---
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
Pytorch Lighting
🚀 Features:
🎨 GradCAM Visualization: User can adjust the visualization opacity and select model layers to be used.
🚫 Misclassified Images: View where the model went wrong.
📷 Image Upload & Examples: User can test prediction on new images which can be uploaded.
🥇 Top Predictions: View model's top class predictions (up to 10).
🔍 How to Use:
📥 Choose an Image: Upload your own or pick from the examples.
🎛️ Tweak Settings: Adjust sliders for GradCAM, top predictions, and misclassified images.
🟢 Submit: Click to get results.
❓ Feedback/Contribution: Got suggestions or wish to help? Reach out!
Usecase: This interface uses a ResNet18 model trained on CIFAR10. GradCAM visualizations highlight the model's focus.
⚠️ Note: Best with CIFAR10-related images.
|