From 1e1bdcb10115089438b67a057084923438c7c716 Mon Sep 17 00:00:00 2001
From: Nikolai Hartmann <Nikolai.Hartmann@physik.uni-muenchen.de>
Date: Fri, 2 Nov 2018 13:18:37 +0100
Subject: [PATCH] option to change default kernel initializer
---
toolkit.py | 19 +++++++++++++++----
1 file changed, 15 insertions(+), 4 deletions(-)
diff --git a/toolkit.py b/toolkit.py
index 46a4885..a0be319 100755
--- a/toolkit.py
+++ b/toolkit.py
@@ -38,6 +38,7 @@ from keras.models import Sequential, Model, model_from_json
from keras.layers import Dense, Dropout, Input, Masking, GRU, LSTM, concatenate, SimpleRNN
from keras.callbacks import History, EarlyStopping, CSVLogger, ModelCheckpoint, TensorBoard
from keras.optimizers import SGD
+import keras.initializers
import keras.optimizers
from keras.utils.vis_utils import model_to_dot
from keras import backend as K
@@ -188,7 +189,9 @@ class ClassificationProject(object):
:param normalize_weights: normalize the weights to mean 1
- :param ignore_neg_weights: ignore events with negative weights in training (default: False)
+ :param ignore_neg_weights: ignore events with negative weights in training - not recommended! (default: False)
+
+ :param kernel_initializer: weight initializer for the dense layers - if None (default) the keras defaults are used
"""
@@ -260,7 +263,8 @@ class ClassificationProject(object):
mask_value=None,
apply_class_weight=True,
normalize_weights=True,
- ignore_neg_weights=False):
+ ignore_neg_weights=False,
+ kernel_initializer=None):
self.name = name
self.signal_trees = signal_trees
@@ -343,6 +347,7 @@ class ClassificationProject(object):
self.apply_class_weight = apply_class_weight
self.normalize_weights = normalize_weights
self.ignore_neg_weights = ignore_neg_weights
+ self.kernel_initializer = kernel_initializer
self.s_train = None
self.b_train = None
@@ -734,7 +739,10 @@ class ClassificationProject(object):
self.dropout,
self.use_bias,
):
- hidden_layer = Dense(node_count, activation=self.activation_function, use_bias=use_bias)(hidden_layer)
+ extra_opts = dict()
+ if self.kernel_initializer is not None:
+ extra_opts["kernel_initializer"] = getattr(keras.initializers, self.kernel_initializer)()
+ hidden_layer = Dense(node_count, activation=self.activation_function, use_bias=use_bias, **extra_opts)(hidden_layer)
if (dropout_fraction is not None) and (dropout_fraction > 0):
hidden_layer = Dropout(rate=dropout_fraction)(hidden_layer)
@@ -1786,7 +1794,10 @@ class ClassificationProjectRNN(ClassificationProject):
flat_channel = flat_input
combined = concatenate(rnn_channels+[flat_channel])
for node_count, dropout_fraction in zip(self.nodes, self.dropout):
- combined = Dense(node_count, activation=self.activation_function)(combined)
+ extra_opts = dict()
+ if self.kernel_initializer is not None:
+ extra_opts["kernel_initializer"] = getattr(keras.initializers, self.kernel_initializer)()
+ combined = Dense(node_count, activation=self.activation_function, **extra_opts)(combined)
if (dropout_fraction is not None) and (dropout_fraction > 0):
combined = Dropout(rate=dropout_fraction)(combined)
combined = Dense(1, activation=self.activation_function_output)(combined)
--
GitLab