Skip to content

Latest commit

 

History

History
36 lines (22 loc) · 1015 Bytes

File metadata and controls

36 lines (22 loc) · 1015 Bytes

Pytorch Conditional-Flow-Matching(CFM) Tutorial

A simple tutorial of Conditional Flow Matching Models (CFMs, Y. Lipman et. al., 2022) using MNIST dataset.

Prerequisites

(1) Download Pytorch and etcs.

(2) Install dependencies via following command

sh install.sh

Expremental Results

  • Used a RTX-3090 GPU for all implementations.

  • trained on MNIST dataset for 200 epochs

  • ground-truth samples

ground_truth

  • generated samples

generated

References

[1] Neural Ordinary Differential Equations, R. T. Q. Chen et. al., 2018

[2] Denoising Diffusion Probabilistic Models, J. Ho et. al., 2020

[3] Flow Matching for Generative Modeling, Y. Lipman et. al., 2022