Skip to main content
. 2022 Feb 14;12:782988. doi: 10.3389/fonc.2022.782988

Table 2.

Parameters of the deep learning networks.

Mask-RCNN U-Net
Batch size 4 4
Learning rate starting at 10–3, and decrease by one-tenth every thirty epochs in the non-augmented dataset and every ten epochs in the augmented dataset. Starting at 10–3, and when the loss does not decrease after five epochs, the learning rate decreases by one-tenth.
Epoch 110 in the non-augmented dataset and 28 in the augmented dataset 110 in the non-augmented dataset and 28 in the augmented dataset
Input image format RGB images (The size of the input images is not fixed) RGB images (416*416*3)