Skip to main content
EURASIP Journal on Advances in Signal Processing

Table 5 Results comparison of single/double layer structure

From: A hierarchically sampling global sparse transformer in data stream mining for lightweight image restoration

 

Patch size

Channels

Epoch

GPUs

PSNR

Train time (h)

HST-tiny

\(128^{2}\)

32

200

1

29.17

12.6

HST-double

\(128^{2}\)

32

200

1

29.96

22.8

Back to article page