Closed Zessay closed 7 months ago
I in-depth study the codes, and I think the null_cond_embed
in the FiLMTransformer
is used as classifier-free guidance
and the cond_drop_prob
is used as control signal, while the ClassifierFreeSampleModel
doesn't work. Is it right?
Hi! This is absolutely correct that this version of the CFG model is broken. Thanks so much for pointing this out! I'll open a PR right now to fix this issue, will run a few tests to make sure this is the one matching what the paper states.
Closing now, after patch fix. Thanks for pointing this out! Please feel to reopen if needed.
Thanks for your excellent work!
I find it says you adopt
classifier-free guidance
policy to train the diffusion module in the paper, as it shows in the following picture.However, in your codes, I find the
cond_mode
parameter is set whenFinLMTransformer
model is initialized, and won't change in theTrainLoop
. Moreover, theforward
function of theFiLMtranformer
only uses thecond_mode
of the model instance, doesn't use the condition signal in they
.So, I wonder whether the
classifier-free guidance
is used in the training process? Looking forward to your reply!