0

I am trying to do a multi-class classification project with CNN. My issue is getting good accuracy but not predicting well on validation data. I have introduced l2 regularization but it is not generalizing well. Also tried with different l2 regularization values (1e-3, 1e-4) Here is my Accuracy graph and Loss graph. Topology:

model = Sequential()
inputs = keras.Input(shape=(512, 512, 3), name="img")
x = Conv2D(32, kernel_size=(3,3), strides=(1,1), kernel_regularizer=l2(1e-5), padding='same')
x = BatchNormalization()(x)
x1 = Activation('relu')(x)
x2 = Conv2D(32, kernel_size=(3,3), strides=(1,1), kernel_regularizer=l2(1e-5), padding='same'
(x1)
x = BatchNormalization()(x2)
x = Activation('relu')(x2)
x3 = Conv2D(32, kernel_size=(3,3), strides=(1,1),  kernel_regularizer=l2(1e-5), padding='same')
(x)
x = BatchNormalization()(x3)
x = tensorflow.keras.layers.add([x, x1]) # ==> Shortcut
x = Activation('relu')(x)
x4 = Conv2D(64, kernel_size=(3,3), strides=(2,2), kernel_regularizer=l2(1e-5),padding='same')(x)
x = BatchNormalization()(x4)
x = Activation('relu')(x)
x5 = Conv2D(64, kernel_size=(3,3), strides=(1,1), kernel_regularizer=l2(1e-5), padding='same')
(x)
x = BatchNormalization()(x5)
x = Activation('relu')(x)
x6 = Conv2D(64, kernel_size=(3,3), strides=(1,1), kernel_regularizer=l2(1e-5), padding='same')
(x)
x = BatchNormalization()(x6)
x = tensorflow.keras.layers.add([x, x4]) # ==> Shortcut
x = Activation('relu')(x)
x7 = Conv2D(128, kernel_size=(3,3), strides=(2,2), kernel_regularizer=l2(1e-5), padding='same')
(x)
x = BatchNormalization()(x7)
x = Activation('relu')(x)
x8 = Conv2D(128, kernel_size=(3,3), strides=(1,1), kernel_regularizer=l2(1e-5), padding='same') 
(x)
x = BatchNormalization()(x8)
x = Activation('relu')(x)
x9 = Conv2D(128, kernel_size=(3,3), strides=(1,1), kernel_regularizer=l2(1e-5), padding='same') 
(x)
x = BatchNormalization()(x9)
x = tensorflow.keras.layers.add([x, x7]) #
x = Activation('relu')(x)

x10 = Conv2D(256, kernel_size=(3,3), strides=(1,1), kernel_regularizer=l2(1e-5), padding='same') 
(x)
x = BatchNormalization()(x10)
x = Activation('relu')(x)

x11 = Conv2D(256, kernel_size=(3,3) , strides=(1,1), kernel_regularizer=l2(1e-5),padding='same') 
(x)
x = BatchNormalization()(x11)
x = Activation('relu')(x)

x12 = Conv2D(256, kernel_size=(3,3), strides=(1,1), kernel_regularizer=l2(1e-5), padding='same') 
(x)
x = BatchNormalization()(x12)
x = tensorflow.keras.layers.add([x, x10]) #
x = Activation('relu')(x)

x13 = Conv2D(512, kernel_size=(3,3), strides=(1,1), kernel_regularizer=l2(1e-5), padding='same') 
(x)
x = BatchNormalization()(x13)
x = Activation('relu')(x)

x14 = Conv2D(512, kernel_size=(3,3), strides=(1,1),kernel_regularizer=l2(1e-5), padding='same') 
(x)
x = BatchNormalization()(x14)
x = Activation('relu')(x)

x15 = Conv2D(512, kernel_size=(3,3), strides=(1,1), kernel_regularizer=l2(1e-5), padding='same')(x)
x = BatchNormalization()(x15)
x = tensorflow.keras.layers.add([x, x13]) #
x = Activation('relu')(x)
x = Flatten()(Conv2D(1, kernel_size=1, strides=(1,1), kernel_regularizer=l2(1e-5), 
 padding='same')(x))
x = layers.Dropout(0.3)(x)
outputs = Dense(4, activation ='softmax', kernel_initializer ='he_normal')(x) 
model = Model(inputs, outputs)
model.summary()

` I tried with different filters, decreasing/increasing layers. Is this issue is because of overfitting? Any suggestion on what I can improve so that I get smoother curve and good predictions.

Shivani Akotkar
  • 1
  • 2
  • It's overfitting. You should add some Drop layers or use Image Augmentation to expand your dataset – dinhanhx Mar 04 '21 at 07:54
  • Question like this should be closed due to [How should we handle machine learning questions asking for performance improvements?](https://meta.stackoverflow.com/a/404108/13358358) – dinhanhx Mar 04 '21 at 07:55

1 Answers1

0
  • You could try putting dropouts in the Conv2D layers too, that should help with some of the overfitting.
  • Decrease the alpha (learning rate of the optimizers), so that the optima doesn't overshoot.

Should help :)

Apurv Singh
  • 21
  • 3