search

Vchitect / Latte

Latte: Latent Diffusion Transformer for Video Generation.
Apache License 2.0
1.45k stars 147 forks source link

Can you provide the code for DDIM sampler #37

Closed lcwLcw123 closed 1 month ago

lcwLcw123 commented 4 months ago

I try to change the ‘sample_method’ hyperparamter to 'DDIM' in Latte/configs/t2v/t2v_sample.yaml, which make the bad performance of the output image. can you provide some scripts for DDIM sampler, or is that model can not work well when using DDIM sampler?

Sicnerely

maxin-cn commented 4 months ago

I try to change the ‘sample_method’ hyperparamter to 'DDIM' in Latte/configs/t2v/t2v_sample.yaml, which make the bad performance of the output image. can you provide some scripts for DDIM sampler, or is that model can not work well when using DDIM sampler?

Sicnerely

I will provide the DDIM sampling method in the future, but I am not sure when it will be ready. If you already have a handle on DDIM's sampling method, welcome to PR.

I have solved the problem of not being able to generate video when using DDIM (https://github.com/Vchitect/Latte/commit/40c2e40fbb82ca92c93cd85d6790a6760e9fad30).