caffe官网关于fine tuning的一个例子

翻译的链接：http://caffe.berkeleyvision.org/gathered/examples/finetune_flickr_style.html

挑一些主要的链接进行翻译

step1:利用shell命令下载数据集：

 python examples/finetune_flickr_style/assemble_data.py --workers=-1 --images=2000 --seed 831486

此时通过这个命令之后，会将训练/测试数据保存在data/flickr_style文件夹里面

step2:下载训练的model:

./scripts/download_model_binary.py models/bvlc_reference_caffenet.

step3:训练下载好的model

./build/tools/caffe train -solver models/finetune_flickr_style/solver.prototxt -weights models/bvlc_reference_caffenet/bvlc_reference_caffenet.caffemodel -gpu 0

参数说明：

train：用来进行训练

-solver：用来进行超参数训练，调用solver.prototxt文件 

-weights 用来对于刚才下载好的caffemodel进行继续训练

---------------------------------------------------------------------------------------------------

对于fine tuning的总结：

Fine-tuning can be feasible when training from scratch would not be for lack of time or data. Even in CPU mode each pass through the training set takes ~100 s. GPU fine-tuning is of course faster still and can learn a useful model in minutes or hours instead of days or weeks. Furthermore, note that the model has only trained on < 2,000 instances. Transfer learning a new task like style recognition from the ImageNet pretraining can require much less data than training from scratch.

如果不缺少时间或者数据的话，那么微调会变得可用。GPU的训练速度远远快于cpu,利用imageNet的一个训练好的模型会远远快于从零开始训练模型