迁移学习的两种策略

In practice, very few people train an entire Convolutional Network from scratch (with random initialization), because it is relatively rare to have a dataset of sufficient size. Instead, it is common to pretrain a ConvNet on a very large dataset (e.g. ImageNet, which contains 1.2 million images with 1000 categories), and then use the ConvNet either as an initialization or a fixed feature extractor for the task of interest.

在实践中，很少有人从头开始训练整个卷积网络(使用随机初始化)，因为拥有足够大小的数据集的情况相对较少。相反，通常是在一个非常大的数据集(例如ImageNet，它包含120万幅图像，1000个类别)上预训练卷积神经网络，然后使用卷积神经网络作为初始化或固定的特征提取器来完成感兴趣的任务。

Finetuning the convnet: Instead of random initializaion, we initialize the network with a pretrained network, like the one that is trained on imagenet 1000 dataset. Rest of the training looks as usual.

优化卷积网络:与随机初始化不同，我们使用预先训练好的网络来初始化网络，就像在imagenet 1000 dataset上训练的那样。其余的训练看起来和往常一样。

ConvNet as fixed feature extractor: Here, we will freeze the weights for all of the network except that of the final fully connected layer. This last fully connected layer is replaced with a new one with random weights and only this layer is trained.

 

作为固定的特征提取器:在这里，我们将冻结所有网络的权重，除了最后的完全连接层。最后的全连接层被一个新的随机权值的层所取代，并且只训练这一层。