Bearing Remaining Useful Life Prediction Based on AdCNN and CWGAN under Few Samples

Algorithm 2

CWGAN training process.

(1)	Initialize: discriminator with parameter , generator with parameter .
(2)	for training iterations do
(3)	for iterations do
(4)	sample m example from dataset
(5)	sample m noise samples from the prior
(6)	obtaining generator data ,
(7)	update by descending along its gradient
(8)
(9)	end for
(10)	for iterations do
(11)	update by descending along its gradient
(12)
(13)	end for
(14)	end for