Implement DDIM with the "trailing" timestep spacing and TCD #568
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…eter for DDIM and TCD, minor comment clarification
yslai
changed the title
stduhpf
reviewed
Jan 14, 2025
examples/cli/main.cpp
Outdated
| @@ -219,7 +224,7 @@ void print_usage(int argc, const char* argv[]) { | |||
| printf(" 1.0 corresponds to full destruction of information in init image\n"); | |||
| printf(" -H, --height H image height, in pixel space (default: 512)\n"); | |||
| printf(" -W, --width W image width, in pixel space (default: 512)\n"); | |||
| printf(" --sampling-method {euler, euler_a, heun, dpm2, dpm++2s_a, dpm++2m, dpm++2mv2, ipndm, ipndm_v, lcm}\n"); | |||
| printf(" --sampling-method {euler, euler_a, heun, dpm2, dpm++2s_a, dpm++2m, dpm++2mv2, ipndm, ipndm_v, lcm, ddim_trailing}\n"); | |||
There was a problem hiding this comment.
You forgot to add TCD here. At some point it will be cleaner to iterate over sample_method_str[] for printing the help message.
…ent reference style
|
Both DDIM and TCD work very well with PCM LoRAs, too. Tested with SD 1.5 and SDXL. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I noticed that due to the structure of k-diffusion, stable-diffusion.cpp does not support DDIM and TCD, which are particularly useful for acceleration techniques like Hyper-SD. This patch implements DDIM ("trailing" timestep spacing version) and TCD on top of the k-diffusion structure, by recalculating the beta_t schedule and cumulative alpha_t products, and then use them to undo the scaling functions in k-diffusion to match DDIM/TCD. The (stochastic) eta parameter in DDIM and TCD is also implemented, in order for TCD to be useful for Hyper-SD in multi-step settings.