EURASIP Journal on Audio, Speech, and Music Processing

Table 3 Training parameters used on the datasets. Bs: batch size, lr: learning rate, wl: warm-up length in epochs, N_e: number of epochs, α: mixup Beta param

From: Comparison of semi-supervised deep learning algorithms for audio classification

	bs	lr	wl	N_e	α
Supervised	256	0.001	-	300	-
Mixup	256	0.001	-	300	0.40
MT	64	0.001	50	200	-
MT+mixup	64	0.001	50	200	0.40
DCT	64	0.0005	160	300	-
DCT+mixup	64	0.0005	160	300	0.40
MM-mixup	256	0.001	-	300	-
MM	256	0.001	-	300	0.75
RMM-mixup	256	0.001	-	300	-
RMM	256	0.001	-	300	0.75
FM	256	0.001	-	300	-
FM+mixup	256	0.001	-	300	0.75

Back to article page