This repository contains the code and trained models of "Sharing Residual Units Through Collective Tensor Factorization in Deep Neural Networks".
| Method | Settings |
|---|---|
| Random Mirror | True |
| Random Crop | 8% - 100% |
| Aspect Ratio | 3/4 - 4/3 |
| Random HSL | [20,40,50] |
Note: We did not use PCA Lighting and any other advanced augmentation methods.
The augmented input images are substrated by mean RGB = [ 124, 117, 104 ], and then multiplied by 0.0167.
Single crop validation error (center 224x224 crop from resized image with shorter side=256):
| Model | Setting | Model Size | Top-1 |
|---|---|---|---|
| CRU-Net-56 @x14 | 32x4d | 98MB | 21.9% |
| CRU-Net-56 @x14 | 136x1d | 98MB | 21.7% |
| CRU-Net-116 @x28x14 | 32x4d | 168MB | 20.6% |
| CRU-Net-116, wider @x28x14 | 64x4d | 318MB | 20.3% |
We also trained a tiny CRU-Net-56 with less than half the size of ResNet-50.
Single crop validation error (center 224x224 crop from resized image with shorter side=256):
| Model | Setting | Model Size | Top-1 |
|---|---|---|---|
| CRU-Net-56,tiny @x14 | 32x4d | 48MB | 22.9% |
10-crop validation accuracy (averaging softmax scores of 10 224x224 crops from resized image with shorter side=256):
| Model | Setting | Model Size | Top-1 |
|---|---|---|---|
| CRU-Net-116 @x28x14 | 32x4d | 163MB | 56.6% |
| Model | Setting | Dataset | Link |
|---|---|---|---|
| CRU-Net-56,tiny @x14 | 32x4d | ImageNet-1k | GoogleDrive |
| CRU-Net-56 @x14 | 32x4d | ImageNet-1k | GoogleDrive |
| CRU-Net-56 @x14 | 136x1d | ImageNet-1k | GoogleDrive |
| CRU-Net-116 @x28x14 | 32x4d | ImageNet-1k | GoogleDrive |
| CRU-Net-116 @x28x14 | 32x4d | Places365-Standard | GoogleDrive |