Synthetic Digit Generator

Demo

Generate a grid of synthetic digits and explore how the output morphs.

• Use the blue “Number” control to choose Auto or a digit from 0–9.
• Click “Generate” to refresh the grid.
• Open “Extra settings” to adjust seed, dim, grid density, minimum grid height, and distortion level.
• Generate again to compare how settings change the outputs.

STAR Summary

• Trained a Variational Autoencoder (VAE) on MNIST (60,000 training digits) with a 20-dim latent space for up to 100 epochs.
• Generated new digits by sampling the learned latent space.
• Saved the trained model so generation is a quick inference step.

Project Links

• GitHub
• Live Demo
• PDF
• Notebook

Notes

Trained on MNIST; generated samples reflect that digit domain rather than real handwriting diversity.

Other Projects

• Smart Sentence Retriever NLP Embeddings & Serverless Retrieval
• Chatbot (LoRA + RAG) RAG Chatbot Fine-Tuned with LoRA
• Shape Classifier Demo Handwritten Shape Recognition