Commit 30791328 authored by Martin Řepa's avatar Martin Řepa

Correct comments about loss weight

parent 33e5c1eb
......@@ -19,9 +19,12 @@ class NeuralNetworkConfig:
# List of available optimizers: https://keras.io/optimizers/
optimizer: str = attr.ib(default='adam')
# Weight of false positive result for training phase
# The bigger it is, less false positives will neural network produce
# Setting it to 1 makes all results neutral
# From docs:
# Value used for weighting the loss function (during training only) for
# malicious requests. This can be useful to tell the model to "pay more
# attention" to malicious samples.
# Setting it to 1 makes loss function behave equally for both predictions
# during training
fp_weight: int = attr.ib(default=5)
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment