Skip to content

Includes mutable collections in the denoiser's forward pass. #84

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
copybara-service wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Includes mutable collections in the denoiser's forward pass. #84

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 4 additions & 3 deletions swirl_dynamics/projects/probabilistic_diffusion/models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -133,17 +133,18 @@ def loss_fn(
vmapped_mult = jax.vmap(jnp.multiply, in_axes=(0, 0))
noised = batch["x"] + vmapped_mult(noise, sigma)
cond = batch["cond"] if self.cond_shape else None
denoised = self.denoiser.apply(
{"params": params},
denoised, updated_mutables = self.denoiser.apply(
{"params": params, **mutables},
x=noised,
sigma=sigma,
cond=cond,
is_training=True,
mutable=mutables.keys() if mutables else False,
rngs={"dropout": rng3}, # TODO: refactor this.
)
loss = jnp.mean(vmapped_mult(weights, jnp.square(denoised - batch["x"])))
metric = dict(loss=loss)
return loss, (metric, mutables)
return loss, (metric, updated_mutables)

def eval_fn(
self, variables: models.PyTree, batch: models.BatchType, rng: Array
Expand Down
4 changes: 4 additions & 0 deletions swirl_dynamics/projects/probabilistic_diffusion/trainers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -116,6 +116,10 @@ def inference_fn_from_state_dict(
variables = state.ema_variables
else:
variables = state.model_variables
if state.flax_mutables:
variables = flax.core.FrozenDict(
{**variables, **state.flax_mutables}
)
return models.DenoisingModel.inference_fn(variables, *args, **kwargs)


Expand Down