Spaces:
Build error
Build error
A newer version of the Gradio SDK is available:
6.5.0
metadata
title: Motion Encoder Decoder
emoji: 🌍
colorFrom: gray
colorTo: blue
sdk: gradio
sdk_version: 5.29.0
app_file: app.py
pinned: false
license: mit
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
🧠 Motion Encoder Decoder ML Pipeline
An interactive Gradio-based machine learning pipeline for generating, training, and testing encoder-decoder models on simulated physics trajectories.
This PyTorch-based system models motion dynamics such as projectile paths and bouncing objects using a sequence-to-sequence architecture.
Features
- ⚙️ Dataset generation (custom physics simulations)
- 🧪 Training with optional early stopping
- 📈 Input sensitivity testing
- 🔮 Real-time predictions and trajectory visualizations
- 📤 Upload / 📥 Download of models and datasets (in
/tmp)
Try It Out
- Select a physics type
- Generate or upload a dataset
- Train a model or upload a pretrained
.pth - Visualize predictions from dynamic input sliders
Built With
- Python 3.13
- PyTorch
- Gradio
- Matplotlib
- NumPy
👨💻 Developed by Miles Exner