Convolutional Neural Networks
GPU-Efficient Network
Introduced by Lin et al. in Neural Architecture Design for GPU-Efficient NetworksGENets, or GPU-Efficient Networks, are a family of efficient models found through neural architecture search. The search occurs over several types of convolutional block, which include depth-wise convolutions, batch normalization, ReLU, and an inverted bottleneck structure.
Source: Neural Architecture Design for GPU-Efficient Networks
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Action Recognition | 1 | 33.33% |
| Ensemble Learning | 1 | 33.33% |
| Object Detection | 1 | 33.33% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
Batch Normalization
|
Normalization | |
Depthwise Convolution
|
Convolutions | |
Inverted Residual Block
|
Skip Connection Blocks | |
ReLU
|
Activation Functions | |
Residual Connection
|
Skip Connections |
