Skip to content

Replace CC tanh activation with swish #901

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 2 commits into from
Jun 22, 2018
Merged

Replace CC tanh activation with swish #901

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
89b5c07
Replace tanh on cc models w/ swish
awjuliani Jun 21, 2018
6f81f65
Change lambda
awjuliani Jun 21, 2018
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
3 changes: 2 additions & 1 deletion python/trainer_config.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -114,10 +114,11 @@ VisualPyramidBrain:

Ball3DBrain:
normalize: true
batch_size: 1200
batch_size: 64
buffer_size: 12000
summary_freq: 1000
time_horizon: 1000
lambd: 0.99
gamma: 0.995
beta: 0.001

Expand Down
5 changes: 1 addition & 4 deletions python/unitytrainers/models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -168,10 +168,7 @@ def create_observation_streams(self, num_streams, h_size, num_layers):
:return: List of encoded streams.
"""
brain = self.brain
if brain.vector_action_space_type == "continuous":
activation_fn = tf.nn.tanh
else:
activation_fn = self.swish
activation_fn = self.swish

self.visual_in = []
for i in range(brain.number_visual_observations):
Expand Down