validation loss and accuracy both are fluctuating.training accuracy and loss have saturated

Question

I am trying to train a classifier using mobilnetv2 with 2 classes.I have freezed the base layers and adding the below lines.

 headModel= baseModel.output
 headModel = AveragePooling2D(pool_size=(7, 7))(headModel)
 headModel = Flatten(name="flatten")(headModel)
 headModel = Dense(128, activation="relu")(headModel)
 headModel = Dropout(0.5)(headModel)
 headModel = Dense(2, activation="softmax")(headModel)