Generative Models

implementation codes for below models with Tensorflow 2.0

1. Generative Modeling by Estimating Gradients of the Data Distribution

Reproduction of "Generative Modeling by Estimating Gradients of the Data Distribution" by Yang Song and Stefano Ermon (NeurIPS 2019) in Tensorflow 2.0. Implementation codes are written with reference to the following github repositories:

• https://github.com/ermongroup/ncsn
• https://github.com/Xemnas0/NCSN-TF2.0

2. Denoising Diffusion Probabilistic Models

Reproduction of "Denoising Diffusion Probabilistic Models" in Tensorflow 2.0. Implementation codes are written with reference to the following github repositories:

• https://github.com/hojonathanho/diffusion

result

• mnist

• cifar10

3. Score-Based Generative Modeling through Stochastic Differential Equations

Reproduction of "Score-Based Generative Modeling through Stochastic Differential Equations" in Tensorflow 2.0. Implementation codes are written with reference to the following github repositories:

• https://github.com/yang-song/score_sde