SqueezeNext: Hardware-Aware Neural Network Design
SqueezeNext: 针对硬件特性的神经网络设计.
SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
SqueezeNet: 与AlexNet精度相当的轻量级模型.
Searching for MobileNetV3
使用神经结构搜索寻找MobileNet V3.
MobileNetV2: Inverted Residuals and Linear Bottlenecks
MobileNetV2: 倒残差与线性瓶颈.
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
MobileNet: 使用深度可分离卷积构造轻量网络.
EfficientNetV2: Smaller Models and Faster Training
EfficientNetV2: 更小的模型和更快的训练.