Table 1 Lists of deep learning architectures selected for the validation and related characteristics: layers used and total number of features extracted.
From: Deep-Manager: a versatile tool for optimal feature selection in live-cell imaging analysis
ResNET101 He et al., 2016 | VGG19 A Bhandary et al., 2020 | NasNETLarge Zoph et al., 2018 | DenseNET201 Huang et al, 2017 | |
|---|---|---|---|---|
Layer | ‘average-pool5’ | ‘max-pool5’ | ‘global_average_pool5’ | ‘average_pool5’ |
N. of features | 2048 | 25088 | 4032 | 1920 |
Input Layer size | 224 × 224 × 3 | 224 × 224 × 3 | 331 × 331 × 3 | 224 × 224 × 3 |
