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

Fix DFPTraining validation set option #709

Merged
merged 3 commits into from
Feb 23, 2023
Merged

Fix DFPTraining validation set option #709

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
7bc0073
fix validation and call to fit
efajardo-nv Feb 14, 2023
7fb30e3
Merge branch 'branch-23.03' into dfp-validation-set-fix
efajardo-nv Feb 22, 2023
4a572e9
Merge branch 'branch-23.03' into dfp-validation-set-fix
efajardo-nv Feb 23, 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
6 changes: 4 additions & 2 deletions examples/digital_fingerprinting/production/morpheus/dfp/stages/dfp_training.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,7 +56,7 @@ def __init__(self, c: Config, model_kwargs: dict = None, epochs=30, validation_s

self._epochs = epochs

if (validation_size > 0.0 and validation_size < 1.0):
if (validation_size >= 0.0 and validation_size < 1.0):
self._validation_size = validation_size
else:
raise ValueError("validation_size={0} should be a positive float in the "
Expand Down Expand Up @@ -85,13 +85,15 @@ def on_data(self, message: MultiDFPMessage):
# Only train on the feature columns
train_df = train_df[train_df.columns.intersection(self._config.ae.feature_columns)]
validation_df = None
run_validation = False

# Split into training and validation sets
if self._validation_size > 0.0:
train_df, validation_df = train_test_split(train_df, test_size=self._validation_size, shuffle=False)
run_validation = True

logger.debug("Training AE model for user: '%s'...", user_id)
model.fit(train_df, epochs=self._epochs, val=validation_df)
model.fit(train_df, epochs=self._epochs, val=validation_df, run_validation=run_validation)
logger.debug("Training AE model for user: '%s'... Complete.", user_id)

output_message = MultiAEMessage(message.meta,
Expand Down