1, 图片只能转换成caffe可读的leveldb或者lmdb存储。
The backend {lmdb, leveldb} for storing the result.
2, How to Setup Caffe to Use Other Datasets by Modifying imagenet Example
http://drubiano.github.io/2014/06/18/caffe-custom-data.html
3, The key function for compute(They are defined in /usr/include/cblas.h, they are Linear Algebra PACKage.):
cpu:
cblas_dgemm, cblas_sgemv, cblas_saxpy, cblas_daxpby
cuda:
cublas_Dgemm, cublasDgemv, cublasSgemm, cublasSgemv, cublasSaxpy, cublasDaxpy, cublasSscal
4, caffe rnn and lstm explanation:
http://stackoverflow.com/questions/32225388/lstm-module-for-caffe
http://tutorial.caffe.berkeleyvision.org/caffe-cvpr15-sequences.pdf
The backend {lmdb, leveldb} for storing the result.
2, How to Setup Caffe to Use Other Datasets by Modifying imagenet Example
http://drubiano.github.io/2014/06/18/caffe-custom-data.html
3, The key function for compute(They are defined in /usr/include/cblas.h, they are Linear Algebra PACKage.):
cpu:
cblas_dgemm, cblas_sgemv, cblas_saxpy, cblas_daxpby
cuda:
cublas_Dgemm, cublasDgemv, cublasSgemm, cublasSgemv, cublasSaxpy, cublasDaxpy, cublasSscal
4, caffe rnn and lstm explanation:
http://stackoverflow.com/questions/32225388/lstm-module-for-caffe
http://tutorial.caffe.berkeleyvision.org/caffe-cvpr15-sequences.pdf
https://github.com/BVLC/caffe/pull/2033
Note: I will continually updated.
扫一扫
345
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
