Skip to content
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

Jump to bottom

Add support for parallel studyjobs #404

Merged
merged 5 commits into from
Feb 27, 2019
Merged

Add support for parallel studyjobs #404

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
25754a2
Add support for parallel studyjobs
Feb 21, 2019
b8b27fb
fix a typo
Feb 22, 2019
51080f0
Reorganize the program a little bit
Feb 26, 2019
be3332b
fix a typo
Feb 26, 2019
ca0e9d4
fix a typo
Feb 26, 2019
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
11 changes: 5 additions & 6 deletions pkg/suggestion/NAS_Reinforcement_Learning/Controller.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,10 +27,11 @@ def __init__(self,
num_replicas=1,
skip_target=0.4,
skip_weight=0.8,
name="controller"):
name="controller",
logger=None):

print("-" * 80)
print("Building Controller")
self.logger = logger
self.logger.info("Building Controller")

self.num_layers = num_layers
self.num_operations = num_operations
Expand Down Expand Up @@ -86,8 +87,7 @@ def _create_params(self):
def _build_sampler(self):
"""Build the sampler ops and the log_prob ops."""

print("-" * 80)
print("Building Controller Sampler")
self.logger.info("Building Controller Sampler")
anchors = []
anchors_w_1 = []

Expand Down Expand Up @@ -201,7 +201,6 @@ def build_trainer(self):

self.train_step = tf.Variable(0, dtype=tf.int32, trainable=False, name=self.name + "_train_step")
tf_variables = [var for var in tf.trainable_variables() if var.name.startswith(self.name)]
print("-" * 80)

self.train_op, self.lr, self.grad_norm, self.optimizer = get_train_ops(
self.loss,
Expand Down
7 changes: 3 additions & 4 deletions pkg/suggestion/NAS_Reinforcement_Learning/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -122,7 +122,6 @@ This neural architecture can be visualized as
![a neural netowrk architecure example](example.png)

## To Do
1. Add support for multiple studyjobs
2. Add support for multiple trials
3. Change LSTM cell from self defined functions in LSTM.py to `tf.nn.rnn_cell.LSTMCell`
4. Store the suggestion checkpoint to PVC in case of nasrl service pod restarts
1. Add support for multiple trials
2. Change LSTM cell from self defined functions in LSTM.py to `tf.nn.rnn_cell.LSTMCell`
3. Store the suggestion checkpoint to PVC to protect against unexpected nasrl service pod restarts
Loading