AmoebaNet is a convolutional neural network found through regularized evolution architecture search. The search space is NASNet, which specifies a space of image classifiers with a fixed outer structure: a feed-forward stack of Inception-like modules called cells. The discovered architecture is shown to the right.
Source: Regularized Evolution for Image Classifier Architecture SearchPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Classification | 4 | 21.05% |
Object Detection | 4 | 21.05% |
Keypoint Detection | 2 | 10.53% |
Semantic Segmentation | 2 | 10.53% |
Image Augmentation | 1 | 5.26% |
Robust Object Detection | 1 | 5.26% |
Real-Time Object Detection | 1 | 5.26% |
Fine-Grained Image Classification | 1 | 5.26% |
Machine Translation | 1 | 5.26% |
Component | Type |
|
---|---|---|
Average Pooling
|
Pooling Operations | |
Convolution
|
Convolutions | |
Max Pooling
|
Pooling Operations | |
Softmax
|
Output Functions | |
Spatially Separable Convolution
|
Convolutions |