* add crop and reduction layer
* caffe2fluid: support crop and reduction layer convertion

* Continuous Evaluation for image classification.

* add 4card kpis

* update batch_size

* caffe2fluid: fix bug in leaky-relu and add interface for prelu convertion
* caffe2fluid:fix code style problem in graph.py

* Add MKL-DNN Benchmarking to CRNN-CTC

* Make crnn-ctc scripts more portable

* Add documentation for cycle to crnn-ctc-reader

* Update crnn-ctc readme for CPU execution

* Merge CRNN-CTC train & inference scripts

* Fix mnist model & ce, kaffe graph yapf issues

* Remove LD_LIBRARY_PATH from crnn-ctc scripts

* CRNN-CTC scripts: set parallel to true

Abort script if batch_size is lower than num of cores

* CRNN-CTC scripts: limit mode options in infer

* CRNN-CTC scripts: set mkldnn parallel to False

* Add more model configure, such as AlexNet, VGG, GoogleNet, DPN, ResNet etc.
* Add download_imagenet2012.sh
* Update doc.
* Add Chinese doc.

* caffe2fluid: fix bug in transforming Scale layer without bias parameter

* caffe2fluid: fix bug about Scale layer transformation and model saving

support multiple data layer;
fix bug about caffe_pb2;

* revise se_resnext for imagenet classification

* Add the method to train a SE-ResNeXt model

The current code can successfully implement the result of SE-ResNeXt-50.

* Update ImageNet2012 URL

* update

* update readme

* Update readme with download URL

* add unzip

* update se_resnext.py with parallel_exe

* delete train function in se_resnext.py

* add eval.py and infer.py

* move cosine_decay from se_resnext.py to train.py

caffe2fluid: support Reshape and Scale convertion; fix bug in dropout; add more tools for comparing results (#880)

* fix code style problems

* fix bug when loading fluid model

* fix code style problem in brain.py

* update readme.md for caffe2fluid

* fix code style and add save_inference_model in caffe2fluid

* fix code style problems

* fix bug when loading fluid model

* fix code style problem in brain.py

* Refine SE-ResNeXt model and use ParallelExecutor.

* Fix code style.

* Follow comments.

2, add examples for imagenet and mnist
3, support new operator: dropout, lrn, concat
4, compatible with old versions of caffe
5, fix bugs