Skip to main content
. Author manuscript; available in PMC: 2021 Nov 10.
Published in final edited form as: Stat Interface. 2019 Mar 11;12(2):283–293. doi: 10.4310/sii.2019.v12.n2.a9

Table 5.

Error rate on TMA images of different sizes by deep learning and RF.

Image size Deep neural network RF
16 × 16 34.92% 32.84%
32 × 32 36.49% 29.56%
64 × 64 35.20% 28.25%
128 × 128 35.89% 28.82%
256 × 256 36.71% 29.70%