We lifted 6 functions out of this paper's own repositories and ran 3 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| hikvision-research/SparseShiftLayer | canonical | 2 of 4 |
| Eunhui-Kim/SSPQ | pwc_unofficial | 1 of 2 |
| Function | Status | Where it lives |
|---|---|---|
| get_dorefa | Ran | Eunhui-Kim/SSPQ/dorefa.py code served (permissive licence) · get_code("dc614005e85656ec") |
| shear_x | Ran | hikvision-research/SparseShiftLayer/models/auto_augment.py code served (permissive licence) · get_code("53613a740588d619") |
| transform_matrix_offset_center | Ran | hikvision-research/SparseShiftLayer/models/auto_augment.py code served (permissive licence) · get_code("3e773c6c5bc796d4") |
| apply_policy | Not yet run | hikvision-research/SparseShiftLayer/models/auto_augment.py code served (permissive licence) · get_code("bdffd25e53ddbbaf") |
| get_logger | Not yet run | hikvision-research/SparseShiftLayer/utils.py code served (permissive licence) · get_code("014d87639fcaa13f") |
| ternarize | Not yet run | Eunhui-Kim/SSPQ/dorefa.py code served (permissive licence) · get_code("6dc98ff502f4a19c") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Shift operation is an efficient alternative over depthwise separable convolution. However, it is still bottlenecked by its implementation manner, namely memory movement. To put this direction forward, a new and novel basic component named Sparse Shift Layer (SSL) is introduced in this paper to construct efficient convolutional neural networks. In this family of architectures, the basic block is only composed by 1x1 convolutional layers with only a few shift operations applied to the intermediate feature maps. To make this idea feasible, we introduce shift operation penalty during optimization and further propose a quantization-aware shift learning method to impose the learned displacement more friendly for inference. Extensive ablation studies indicate that only a few shift operations are sufficient to provide spatial information communication. Furthermore, to maximize the role of SSL, we redesign an improved network architecture to Fully Exploit the limited capacity of neural Network (FE-Net). Equipped with SSL, this network can achieve 75.0% top-1 accuracy on ImageNet with only 563M M-Adds. It surpasses other counterparts constructed by depthwise separable convolution and the networks searched by NAS in terms of accuracy and practical speed.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("1903.05285")
get_code_for_paper("1903.05285")
have("1903.05285")
Connect an agent — have() is free.