ml-recall

#inverted-dropout

2 questions from Training Deep Networks, PyTorch

Timed test

2 questions · 2 min · score and explanations at the end

Take a test

Q 1 / 2Score 0 / 0

Shuffle questions

Question 1: Short answer

mediumAI-checkedTraining Deep Networks

nn.Dropout(p=0.1) is followed directly by nn.Dropout(p=0.2). Both use inverted dropout, as in PyTorch: an element is zeroed with probability $p$ and otherwise scaled by $1 / (1 - p)$ . The pair behaves exactly like a single nn.Dropout(p). What is $p$ ? Give 2 decimals.

Enter: check / next