Decomposition Specification Reference of Open-Source Network Models
The accuracy specification is top1 ACC(%) for an image classification network, mAP(%) for an object detection network, or DSC(%) for an image segmentation network. The fine-tuning learning rate decreases from 0.1 times of the original learning rate.
Model |
Task Type |
Dataset |
Baseline Accuracy |
Accuracy After Decomposition |
Accuracy After Post-decomposition Fine-tuning |
|---|---|---|---|---|---|
ResNet-18 |
Classification |
ImageNet |
70.66 |
44.02 |
70.34 |
ResNet-34 |
Classification |
ImageNet |
74.2 |
54.92 |
74.15 |
ResNet-50 |
Classification |
ImageNet |
75.6 |
73.64 |
75.91 |
ResNet-101 |
Classification |
ImageNet |
78.52 |
76.97 |
78.24 |
Inception-v3 |
Classification |
ImageNet |
77.98 |
76.95 |
77.78 |
SSD |
Detection |
COCO 2017 |
27.2 |
24.2 |
27.9 |
Faster-RCNN |
Detection |
COCO 2017 |
32.5 |
31 |
32.2 |
Mask-RCNN |
Detection |
COCO 2017 |
37.9 |
36.8 |
38 |
U-Net |
Segmentation |
SSTEM |
87.63 |
85.05 |
87.57 |
Parent topic: Reference