文件列表:

CUDA-convnet (0, 2014-06-21)
CUDA-convnet\ABOUT (159, 2014-03-03)
CUDA-convnet\build.sh (1063, 2014-03-03)
CUDA-convnet\code (0, 2014-06-21)
CUDA-convnet\code\.svn (0, 2014-06-21)
CUDA-convnet\code\.svn\entries (3, 2014-06-21)
CUDA-convnet\code\.svn\format (3, 2014-06-21)
CUDA-convnet\code\.svn\pristine (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\08 (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\08\0809c61e2ee21a90c6da3fd422fcf980d1461fa7.svn-base (9316, 2014-06-21)
CUDA-convnet\code\.svn\pristine\0b (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\0b\0b6311ba97ff644521da135d7a9a932c22eff36e.svn-base (3109, 2014-06-21)
CUDA-convnet\code\.svn\pristine\0c (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\0c\0c09331cad5259ac43db268da1987cd18142fe65.svn-base (935, 2014-06-21)
CUDA-convnet\code\.svn\pristine\0d (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\0d\0d133c71f3733dd1773087d6e0dcf7676e4c97fb.svn-base (363, 2014-06-21)
CUDA-convnet\code\.svn\pristine\0f (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\0f\0fe245cdbb4744a1d2d62c7efd30a14a75d23f9f.svn-base (55944, 2014-06-21)
CUDA-convnet\code\.svn\pristine\11 (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\11\118debc199a04bd3cf3e0c264204ff8340868a68.svn-base (88412, 2014-06-21)
CUDA-convnet\code\.svn\pristine\14 (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\14\143f481860ab39dad1001c0904b2c3476775ffc7.svn-base (14245, 2014-06-21)
CUDA-convnet\code\.svn\pristine\14\146fb2018d1cb7258516d962199d2e108e08c69d.svn-base (14942, 2014-06-21)
CUDA-convnet\code\.svn\pristine\14\148f5ec3702f8ac3c9efc41df70e60cba68f87f6.svn-base (221849, 2014-06-21)
CUDA-convnet\code\.svn\pristine\14\14de96b934a32624d0921eed1d0f06767959bbcd.svn-base (11637, 2014-06-21)
CUDA-convnet\code\.svn\pristine\15 (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\15\15e88488c9dcff78c17a8a016f96be879394e50f.svn-base (17504, 2014-06-21)
CUDA-convnet\code\.svn\pristine\16 (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\16\1661aae0b4ece718e5c0c8d0fbd4cfe8180e769c.svn-base (1325, 2014-06-21)
CUDA-convnet\code\.svn\pristine\18 (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\18\1847c93e1d2dda4acabb521a976513298dd8985c.svn-base (2877, 2014-06-21)
CUDA-convnet\code\.svn\pristine\18\188dc2e40c404718f34b22cfb105a4de0926a383.svn-base (21586, 2014-06-21)
CUDA-convnet\code\.svn\pristine\19 (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\19\1902db0c32795952d22d3d8397f6c6fd9b355c27.svn-base (11419, 2014-06-21)
CUDA-convnet\code\.svn\pristine\19\195f4ff15440695b19c9823ed92a7347b84cb0eb.svn-base (1456, 2014-06-21)
CUDA-convnet\code\.svn\pristine\1a (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\1a\1afdd736eed21bf2287aced477163d4a8ccf4fb1.svn-base (16477, 2014-06-21)
CUDA-convnet\code\.svn\pristine\1c (0, 2014-06-21)
CUDA-convnet\code\.svn\pristine\1c\1c037b75648d9e3a90e0a075dd6a594a61474993.svn-base (7739, 2014-06-21)
CUDA-convnet\code\.svn\pristine\1e (0, 2014-06-21)
... ...

This is my fork of the ``cuda-convnet`` convolutional neural network implementation written by Alex Krizhevsky. ``cuda-convnet`` has quite extensive documentation itself. Find the `MAIN DOCUMENTATION HERE `_. =================== Additional features =================== This document will only describe the small differences between ``cuda-convnet`` as hosted on Google Code and this version. Dropout ======= Dropout is a relatively new regularization technique for neural networks. See the `Improving neural networks by preventing co-adaptation of feature detectors `_ and `Improving Neural Networks with Dropout `_ papers for details. To set a dropout rate for one of our layers, we use the ``dropout`` parameter in our model's ``layer-params`` configuration file. For example, we could use dropout for the last layer in the CIFAR example by modifying the section for the fc10 layer to look like so:: [fc10] epsW=0.001 epsB=0.002 # ... dropout=0.5 In practice, you'll probably also want to double the number of ``outputs`` in that layer. CURAND random seeding ===================== An environment variable ``CONVNET_RANDOM_SEED``, if set, will be used to set the CURAND library's random seed. This is important in order to get reproducable results. Updated to work with CUDA via CMake =================================== The build configuration and code has been updated to work with CUDA via CMake. Run ``cmake .`` and then ``make``. If you have an alternative BLAS library just set it with for example ``cmake -DBLAS_LIBRARIES=/usr/lib/libcblas.so .``.

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：