2016 ilsvrc
2019.08.31 21:20
Trimps-Soushen used the pretrained models from Inception-v3, Inception-v4, Inception-ResNet-v2, Pre-Activation ResNet-200, and Wide ResNet (WRN-68–2) for classification, and found out Top-10 difficult categories as above. Diverse results are obtained, which means there is no models being dominant for all categories. Each of the models are strong at classifying some categories, but also weak at classifying some categories. The diversity of models can be used for improving the accuracy. During training, Trimps-Soushen just performed multi-scale augmentation & large mini batch size. During testing, multi-scale + flip are used with dense fusion.
댓글 0
번호 | 제목 | 글쓴이 | 날짜 | 조회 수 |
---|---|---|---|---|
111 | pytorch - caffe | WHRIA | 2019.12.19 | 37 |
110 | kaggle leakage | WHRIA | 2019.12.19 | 35 |
109 | pytorch object detect / retinanet | WHRIA | 2019.12.22 | 214 |
108 | lvm 확장 [1] | WHRIA | 2019.12.25 | 48 |
107 | 윈도우 raid ahci 전환 | WHRIA | 2019.12.26 | 37 |
106 | melanoma awareness [1] | WHRIA | 2019.12.27 | 45 |
105 | R graph | WHRIA | 2019.12.29 | 228 |
104 | 참고 또 참고 | WHRIA | 2020.01.03 | 69 |
103 | add extra raid disk | WHRIA | 2020.01.05 | 80 |
102 | raid 6 rebuild | WHRIA | 2020.01.07 | 35 |
101 | usb 3.1 + DP | WHRIA | 2020.01.07 | 167 |
100 | lvm [2] | WHRIA | 2020.01.09 | 39 |
99 | retinanet nvidia [3] | WHRIA | 2020.01.12 | 68 |
98 | nvidia caffe | WHRIA | 2020.01.12 | 46 |
97 | xml json pascal [4] | WHRIA | 2020.01.12 | 64 |