# resnetgit
https://github.com/tensorflow/models/blob/master/research/slim/nets/resnet_v1.pygithub
https://github.com/tensorflow/models/blob/master/research/slim/nets/resnet_v2.py.net
https://github.com/tensorflow/models/blob/master/research/slim/nets/resnet_utils.pyblog
# mutil gpu train: (loss = softmax + l2_loss)ci
https://github.com/tensorflow/models/blob/1af55e018eebce03fb61bba9959a04672536107d/tutorials/image/cifar10/cifar10_multi_gpu_train.pyit
https://github.com/tensorflow/models/blob/1af55e018eebce03fb61bba9959a04672536107d/tutorials/image/cifar10/cifar10.pyast
# resnet (loss = softmax + l2_loss)tensorflow
https://github.com/AthenaZof/my-ResNet-tutor/blob/2734963b3f89f4e78a6378cf70b4bed9afb00a86/resnet_model.pymodel
https://blog.csdn.net/wayne2019/article/details/78756218im
https://www.jianshu.com/p/243ee5803837
https://zhuanlan.zhihu.com/p/49615848
https://zhuanlan.zhihu.com/p/51278710
[peper] Improved Bottleneck Features Using Pretrained Deep Neural Networks
https://www.semanticscholar.org/paper/Improved-Bottleneck-Features-Using-Pretrained-Deep-Yu-Seltzer/de8d30f9c59be0c235ae2de7da77993e54f9f91f