I want to understand what does dropout layer do
I read a lot about it but I can’t understand the effect of it to the data to prevent overfitting
import tensorflow as tf
inputs = np.arange(10).reshape(5, 2).astype(np.float32)
layer = tf.keras.layers.Dropout(0.2, input_shape=(2, ))
outputs = layer(inputs, training = true)
print(outputs)
when every run of the application, outputs become different
Welcome to The Tensorflow Forum!
when every run of the application, outputs become different
Yes, Dropout layer randomly sets input units to 0 with a frequency of rate at each step during training time, which helps prevent overfitting.
For example, a given layer would normally have returned a vector [0.2, 0.5, 1.3, 0.8, 1.1] for a given input sample during training; after applying dropout, this vector will have a few zero entries distributed at random, e.g. [0, 0.5, 1.3, 0, 1.1] .
Please refer to this example to check dropout role at reducing overfitting.
Thank you!
Dropout is a great technique for preventing overfitting. I analyzed it and tried to explain it in a very clear way in this article.
I hope you find it usefult