Skip to content

Fix TF Nan bug #1178

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
Sep 5, 2018
Merged

Fix TF Nan bug #1178

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
7743c92
Fix for TF NaNs
awjuliani Sep 5, 2018
1775a42
New soccer model
awjuliani Sep 5, 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
Binary file added UnitySDK/Assets/ML-Agents/Examples/Soccer/TFModels/SoccerTwos.bytes
View file
Open in desktop
Binary file not shown.
7 changes: 7 additions & 0 deletions UnitySDK/Assets/ML-Agents/Examples/Soccer/TFModels/SoccerTwos.bytes.meta
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

18 changes: 11 additions & 7 deletions config/trainer_config.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,21 +67,25 @@ BigWallBrain:
normalize: false

StrikerBrain:
max_steps: 1.0e5
max_steps: 5.0e5
learning_rate: 1e-3
batch_size: 128
buffer_size: 2048
beta: 5.0e-3
num_epoch: 3
buffer_size: 2000
beta: 1.0e-2
hidden_units: 256
summary_freq: 2000
time_horizon: 128
num_layers: 2
normalize: false

GoalieBrain:
max_steps: 1.0e5
batch_size: 128
buffer_size: 2048
beta: 5.0e-3
max_steps: 5.0e5
learning_rate: 1e-3
batch_size: 320
num_epoch: 3
buffer_size: 2000
beta: 1.0e-2
hidden_units: 256
summary_freq: 2000
time_horizon: 128
Expand Down
5 changes: 3 additions & 2 deletions ml-agents/mlagents/trainers/models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -153,9 +153,10 @@ def create_discrete_action_masking_layer(all_logits, action_masks, action_size):
action_idx = [0] + list(np.cumsum(action_size))
branches_logits = [all_logits[:, action_idx[i]:action_idx[i + 1]] for i in range(len(action_size))]
branch_masks = [action_masks[:, action_idx[i]:action_idx[i + 1]] for i in range(len(action_size))]
raw_probs = [tf.multiply(tf.nn.softmax(branches_logits[k]), branch_masks[k]) + (1-branch_masks[k])*1.0e-10
raw_probs = [tf.multiply(tf.nn.softmax(branches_logits[k]), branch_masks[k]) + 1.0e-10
for k in range(len(action_size))]
normalized_probs = [tf.divide(raw_probs[k], tf.reduce_sum(raw_probs[k], axis=1, keepdims=True))
normalized_probs = [
tf.divide(raw_probs[k], tf.reduce_sum(raw_probs[k] + 1.0e-10, axis=1, keepdims=True))
for k in range(len(action_size))]
output = tf.concat([tf.multinomial(tf.log(normalized_probs[k]), 1) for k in range(len(action_size))], axis=1)
return output, tf.concat([tf.log(normalized_probs[k]) for k in range(len(action_size))], axis=1)
Expand Down