Skip to content

Add modulator #190

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

Merged
merged 4 commits into from
Mar 3, 2023
Merged

Add modulator #190

Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
3f4baf6
add possibility to subtract behavior mean in NeuroNormalizer
Feb 27, 2023
eb2ae3f
Add modulator
Feb 28, 2023
7902609
Change order of imports for isort
Feb 28, 2023
a64fb3d
Fix dimensions of output array when `n_parameters_to_modulate` is equ…
Mar 1, 2023
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
Prev Previous commit
Fix dimensions of output array when n_parameters_to_modulate is equ… 
…al to 1
  • Loading branch information
kklur committed Mar 1, 2023
commit a64fb3dafd46184eedafa41b9cf5fb9af45f2e56
5 changes: 3 additions & 2 deletions neuralpredictors/layers/modulators/mlp.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,8 +12,9 @@ def __init__(
):
super().__init__()
warnings.warn("Ignoring input {} when creating {}".format(repr(kwargs), self.__class__.__name__))
self.n_parameters_to_modulate = n_parameters_to_modulate
self.modulator_networks = nn.ModuleList()
for _ in range(n_parameters_to_modulate):
for _ in range(self.n_parameters_to_modulate):

prev_output = input_channels
feat = []
Expand All @@ -36,7 +37,7 @@ def forward(self, x, behavior):
for network in self.modulator_networks:
# Make modulation positive. Exponential would result in exploding modulation -> Elu+1
mods.append(nn.functional.elu(network(behavior)) + 1)
mods = torch.stack(mods)
mods = mods[0] if self.n_parameters_to_modulate == 1 else torch.stack(mods)
return x * mods


Expand Down