2.
Lightweight
search
Quick search
code
Show Source
PDF
Github
Dive into cheap deep learning
Table Of Contents
Getting Started
1. Introduction
1.1. time
1.2. 技术
1.3. 隐私
1.4. money
1.5. Data
2. Lightweight
2.1. Lightweight
2.2. SqueezeNet
2.3. MobileNet
2.4. MobileNet-v2
2.5. ShuffleNet
2.6. GhostNet
3. Compression
3.1. 模型压缩
3.2. 参数剪枝(Pruning)
3.3. Knowledge-Distillation
3.4. 量化
4. Write code
4.1. Jupyter
4.2. API
5. Train
5.1. Server
5.2. Active Learning
5.3. Pretrain
5.4. 改进
5.5. 结构
6. Deployment
6.1. 芯片
6.2. Edge
6.3. mobile
6.4. MCU
6.5. AI 中台
Dive into cheap deep learning
Table Of Contents
Getting Started
1. Introduction
1.1. time
1.2. 技术
1.3. 隐私
1.4. money
1.5. Data
2. Lightweight
2.1. Lightweight
2.2. SqueezeNet
2.3. MobileNet
2.4. MobileNet-v2
2.5. ShuffleNet
2.6. GhostNet
3. Compression
3.1. 模型压缩
3.2. 参数剪枝(Pruning)
3.3. Knowledge-Distillation
3.4. 量化
4. Write code
4.1. Jupyter
4.2. API
5. Train
5.1. Server
5.2. Active Learning
5.3. Pretrain
5.4. 改进
5.5. 结构
6. Deployment
6.1. 芯片
6.2. Edge
6.3. mobile
6.4. MCU
6.5. AI 中台
2.
Lightweight
¶
2.1. Lightweight
2.1.1. 怎样才算?
2.1.2. 方法
2.2. SqueezeNet
2.2.1. 动机
2.2.2. Background
2.2.3. 微观结构
2.2.4. 宏观结构
2.2.5. 核心思路
2.3. MobileNet
2.3.1. ResNet
2.3.2. Activation
2.3.3. 轻量化网络的客观需求
2.3.4. 本文方法
2.3.5. 结构
2.3.6. 深度可分离卷积
2.3.7. MoblieNets瘦身[10]
2.4. MobileNet-v2
2.5. ShuffleNet
2.5.1. 动机
2.5.2. 方法
2.5.3. 分组点卷积Group convolutions`
2.5.4. 通道重排(channel shuffle)
2.5.5. 采用concat替换add操作
2.5.6. FLOPS
2.5.7. ShuffleNet-V28
2.5.8. Comparison with MobileNetV16
2.5.9. ShuffleNet-v2具有高精度的原因
2.6. GhostNet
Previous
1.5. Data
Next
2.1. Lightweight