TimeVQVAE
Synthetic time series generation with vector quantization
All Skills in This Repository (6)
Pure Emerald Level Indicators
ML4ITSuv-pypi-publish
Publish Python packages to PyPI using uv with token management.
ML4ITSgitnexus-refresh-on-stale
Refresh stale GitNexus indexes and sync local skills automatically.
ML4ITSgitnexus-debugging
Trace execution flows and dependencies in a codebase using Git history.
ML4ITSgitnexus-exploring
Navigate Git repository code components, execution flows, and symbols via GitNexus knowledge graph.
ML4ITSgitnexus-impact-analysis
Analyze codebase changes to identify affected dependencies and risk levels.
ML4ITSgitnexus-refactoring
Plan code refactorings by analyzing dependencies and blast radii.
Frequently Asked Questions
FAQPage SchemaHow to install TimeVQVAE?▼
Run `npx skills add ML4ITS/TimeVQVAE --all -g -y` in your terminal to install all skills in this suite globally.
What is TimeVQVAE used for?▼
It generates synthetic time series data by compressing signals into discrete tokens with a VQ-VAE, then sampling new sequences with a bidirectional transformer prior.
How to generate synthetic time series with TimeVQVAE?▼
Train the Stage 1 VQVAE on your data, then use the Stage 2 MaskGIT model's iterative decoding to sample new sequences, optionally conditioned on a class label.
Can I try TimeVQVAE without local setup?▼
Yes. A Google Colab notebook is included where you set the dataset name and sample count, then run unconditional or class-conditional generation on a GPU.
Does TimeVQVAE support class-conditional generation?▼
Yes. The MaskGIT prior model accepts class conditions and uses classifier-free guidance to control the category of generated time series.
Related Repositories in Data & Analytics
View All in Data & Analytics→
PaddlePaddlePaddleOCR
Extract text, tables, and formulas from PDFs and images
D4VinciScrapling
Scrape any website and bypass anti-bot protection with AI
mvanhornlast30days-skill
Research any topic across Reddit, X, YouTube, and the web