add regularization to small networks, fix model name in args, fix visualizations

models/pauls_networks.py

@@ -1,11 +1,12 @@
+from collections import namedtuple
+
 import keras
 from keras.engine import Input, Model as KerasModel
-from keras.layers import Embedding, Conv1D, GlobalMaxPooling1D, Dense, Dropout, Activation, TimeDistributed
+from keras.layers import Activation, Conv1D, Dense, Dropout, Embedding, GlobalMaxPooling1D, TimeDistributed
+from keras.regularizers import l2
 
 import dataset
 
-from collections import namedtuple
-
 Model = namedtuple("Model", ["in_domains", "in_flows", "out_client", "out_server"])
 
 best_config = {
@@ -33,10 +34,14 @@ best_config = {
 def get_embedding(embedding_size, input_length, filter_size, kernel_size, hidden_dims, drop_out=0.5) -> KerasModel:
     x = y = Input(shape=(input_length,))
     y = Embedding(input_dim=dataset.get_vocab_size(), output_dim=embedding_size)(y)
-    y = Conv1D(filter_size, kernel_size, activation='relu')(y)
+    y = Conv1D(filter_size,
+               kernel_size,
+               kernel_regularizer=l2(0.01),
+               activation='relu')(y)
     y = GlobalMaxPooling1D()(y)
     y = Dropout(drop_out)(y)
-    y = Dense(hidden_dims)(y)
+    y = Dense(hidden_dims,
+              kernel_regularizer=l2(0.01))(y)
     y = Activation('relu')(y)
     return KerasModel(x, y)
 
@@ -50,12 +55,13 @@ def get_model(cnnDropout, flow_features, domain_features, window_size, domain_le
     # CNN processing a small slides of flow windows
     y = Conv1D(cnn_dims,
                kernel_size,
+               kernel_regularizer=l2(0.01),
                activation='relu',
                input_shape=(window_size, domain_features + flow_features))(merged)
     # remove temporal dimension by global max pooling
     y = GlobalMaxPooling1D()(y)
     y = Dropout(cnnDropout)(y)
-    y = Dense(dense_dim, activation='relu')(y)
+    y = Dense(dense_dim, kernel_regularizer=l2(0.01), activation='relu')(y)
     out_client = Dense(1, activation='sigmoid', name="client")(y)
     out_server = Dense(1, activation='sigmoid', name="server")(y)
 
@@ -68,18 +74,25 @@ def get_new_model(dropout, flow_features, domain_features, window_size, domain_l
     ipt_flows = Input(shape=(window_size, flow_features), name="ipt_flows")
     encoded = TimeDistributed(cnn)(ipt_domains)
     merged = keras.layers.concatenate([encoded, ipt_flows], -1)
-    y = Dense(dense_dim, activation="relu", name="dense_server")(merged)
+    y = Dense(dense_dim,
+              kernel_regularizer=l2(0.01),
+              activation="relu",
+              name="dense_server")(merged)
     out_server = Dense(1, activation="sigmoid", name="server")(y)
     merged = keras.layers.concatenate([merged, y], -1)
     # CNN processing a small slides of flow windows
     y = Conv1D(cnn_dims,
                kernel_size,
+               kernel_regularizer=l2(0.01),
                activation='relu',
                input_shape=(window_size, domain_features + flow_features))(merged)
     # remove temporal dimension by global max pooling
     y = GlobalMaxPooling1D()(y)
     y = Dropout(dropout)(y)
-    y = Dense(dense_dim, activation='relu', name="dense_client")(y)
+    y = Dense(dense_dim,
+              kernel_regularizer=l2(0.01),
+              activation='relu',
+              name="dense_client")(y)
 
     out_client = Dense(1, activation='sigmoid', name="client")(y)