Evan Shelhamer
|
73001f2fc0
|
fix test data layer post-lmdb
|
2014-06-25 17:32:27 -07:00 |
Jonathan L Long
|
aa7371147e
|
turn off some warnings for older compilers
|
2014-06-25 15:11:42 -07:00 |
Jonathan L Long
|
8490718f4c
|
add WARNINGS to CXXFLAGS
|
2014-06-25 15:11:41 -07:00 |
Jonathan L Long
|
ff5e35ee77
|
upgrade warnings to -Wall -Werror -Wno-sign-compare
|
2014-06-25 15:11:41 -07:00 |
Jonathan L Long
|
cd81969b24
|
don't end comments with \, so that -Wcomment can be used
|
2014-06-25 15:11:41 -07:00 |
Jonathan L Long
|
1334c3fa58
|
initialize and comment variables that the compiler finds suspicious
|
2014-06-25 15:11:41 -07:00 |
Jonathan L Long
|
c41c15a138
|
move CUDA 6.0 check into switch statement itself
This allows -Wswitch to be turned on so that the compiler can check
exhaustiveness.
|
2014-06-25 15:11:41 -07:00 |
Jonathan L Long
|
e84dfb26bb
|
add missing const qualifiers to MemoryDataLayer ExactNum* functions
|
2014-06-25 15:11:41 -07:00 |
Jonathan L Long
|
88472c9b17
|
remove unused variables from tests
|
2014-06-25 15:11:41 -07:00 |
Jonathan L Long
|
a65cb967e6
|
remove unused variables
|
2014-06-25 15:11:40 -07:00 |
Jonathan L Long
|
1ee22bf761
|
initialize in declared order in tests
|
2014-06-25 15:11:38 -07:00 |
Jonathan L Long
|
636d51c4a6
|
initialize in declared order
|
2014-06-25 15:09:12 -07:00 |
Jonathan L Long
|
c0bcf68f8d
|
check if window file is empty in WindowDataLayer
Also note that window files containing windows with different numbers of
channels may not work correctly.
|
2014-06-25 15:09:12 -07:00 |
Jonathan L Long
|
ba09a9c348
|
actually check status values from all HDF5 calls
|
2014-06-25 15:09:12 -07:00 |
Evan Shelhamer
|
9b45809435
|
Merge pull request #427 from jamt9000/fix-kernel-index
Should not modify index in im2col kernel loop
|
2014-06-25 14:47:06 -07:00 |
Yangqing Jia
|
71d6429c4d
|
Merge pull request #508 from kloudkl/ImageDataLayer-RNG-core-dump
Image data layer rng core dump
|
2014-06-23 08:56:30 -04:00 |
Evan Shelhamer
|
8e225e4669
|
Merge pull request #529 from crizCraig/patch-3
Fix example text: there are 256 filters in conv2.
|
2014-06-22 12:17:45 +08:00 |
Craig Quiter
|
93049bd7d5
|
There are 256 filters in conv2.
AlexNet paper splits conv2 into two GPU's, each with 128 filters. Reference paper: http://www.cs.toronto.edu/~fritz/absps/imagenet.pdf Applying #528 to dev instead of master per @Yangqing.
|
2014-06-21 15:22:05 -07:00 |
Jonathan L Long
|
312f413800
|
Merge pull request #398 from sguada/L2_hinge_loss
|
2014-06-21 14:17:50 -07:00 |
Jonathan L Long
|
bddd10ae03
|
explicitly name L1 hinge test
|
2014-06-21 14:15:29 -07:00 |
Jonathan L Long
|
8042be6966
|
fix whitespace error in HingeLossLayer
|
2014-06-21 14:15:29 -07:00 |
Sergio
|
60ff8bacdd
|
Change hinge_norm to norm in test_hinge_loss
|
2014-06-20 20:26:51 -07:00 |
Sergio
|
909fb39240
|
Remove C_ mentions, extra spaces and change hinge_norm to norm
|
2014-06-20 20:22:34 -07:00 |
Sergio
|
4c71114054
|
Unify L1 and L2 Hinge_Loss to follow convention
|
2014-06-20 18:39:01 -07:00 |
Sergio
|
4094170e85
|
Fix the loss to follow the convention
Conflicts:
src/caffe/layers/loss_layer.cpp
|
2014-06-20 18:35:25 -07:00 |
Sergio
|
7be676cfb7
|
Fixed switch and test l2hingeloss
Conflicts:
src/caffe/layers/loss_layer.cpp
src/caffe/test/test_hinge_loss_layer.cpp
|
2014-06-20 18:34:24 -07:00 |
Sergio
|
eaa30176f4
|
Remove spaces and merge tests into one file
Conflicts:
src/caffe/layers/loss_layer.cpp
src/caffe/test/test_hinge_loss_layer.cpp
|
2014-06-20 18:32:49 -07:00 |
Sergio
|
f25687e1be
|
Removed L2HingeLoss class now a case within HingeLoss class
Conflicts:
include/caffe/vision_layers.hpp
src/caffe/layers/loss_layer.cpp
src/caffe/proto/caffe.proto
src/caffe/test/test_l2_hinge_loss_layer.cpp
|
2014-06-20 18:29:14 -07:00 |
Sergio
|
f2452d8966
|
Merge HingeLoss and L2HingeLoss by adding hinge_norm to params
Conflicts:
src/caffe/layers/loss_layer.cpp
src/caffe/proto/caffe.proto
src/caffe/test/test_l2_hinge_loss_layer.cpp
|
2014-06-20 18:26:05 -07:00 |
Kai Li
|
ea77c96356
|
Initialize the RNG generator with an orthogonally newed Generator
|
2014-06-20 17:28:07 +08:00 |
Kai Li
|
bb5f9e6752
|
Fix the condition prefetch_needs_rand in the ImageDataLayer
|
2014-06-20 17:28:07 +08:00 |
Jonathan L Long
|
3d0bc1acbd
|
remove erroneous comment in ArgMaxLayer
|
2014-06-20 01:05:41 -07:00 |
Sergey Karayev
|
430c9ba84c
|
Merge pull request #521 from sguada/set_device_id_at_init
Set device_id at the beginning of Solver.Init()
|
2014-06-19 17:13:34 -07:00 |
Sergio
|
3686f872f2
|
Set device_id at the begining of Solver.Init() to avoid using memory in the default GPU
|
2014-06-19 12:55:48 -07:00 |
Yangqing Jia
|
6af51c3492
|
Merge pull request #504 from leelurch/Config-Example-Ubuntu14.04
Add comment for how to set the CUDA path when cuda tools are installed b...
|
2014-06-18 18:18:08 -07:00 |
Sergio Guadarrama
|
0d7aadc00a
|
Merge pull request #507 from longjon/set-device-early
Fix Caffe::SetDevice to avoid initializing on default device
|
2014-06-17 09:20:25 -07:00 |
Jonathan L Long
|
47ef5629d8
|
in Caffe::SetDevice, call cudaSetDevice before Get
Otherwise initialization will be performed on whichever device is
default.
|
2014-06-16 15:20:17 -07:00 |
Sergey Karayev
|
b50f591a76
|
Merge pull request #431 from mavenlin/lmdb
Add support for LMDB (LevelDB alternative)
|
2014-06-15 23:38:42 -07:00 |
leelurch
|
40837161e3
|
Add comment for how to set the CUDA path when cuda tools are installed by the package manager.
|
2014-06-15 20:07:04 -05:00 |
linmin
|
dbc5ebf31b
|
fix string compare error
|
2014-06-14 16:35:48 +08:00 |
linmin
|
01481d581b
|
add lmdb support for compute_image_mean
|
2014-06-14 16:15:44 +08:00 |
linmin
|
e0ce93383e
|
add lmdb support for convert_imageset
|
2014-06-14 16:15:18 +08:00 |
Evan Shelhamer
|
6b6ab3339f
|
Merge pull request #495 from jeffdonahue/refactor-net
Minor Net::Init refactoring: name loop indices, add helpers
|
2014-06-13 13:22:22 -07:00 |
Evan Shelhamer
|
27ece93816
|
add net surgery link to docs (+ drop old comment)
|
2014-06-13 11:42:40 -07:00 |
Jeff Donahue
|
1dd0d94bbe
|
unify data layer tests: was copied four times for all combinations of
cpu/gpu and leveldb/lmdb; now just one copy of each test body
|
2014-06-12 23:15:12 -07:00 |
Jeff Donahue
|
f8ebd09d06
|
unify test_data_layer tests
|
2014-06-12 22:51:33 -07:00 |
Jeff Donahue
|
c17e413b30
|
lint
|
2014-06-12 22:29:59 -07:00 |
linmin
|
c77c6e12e4
|
fixed cpplint error
|
2014-06-12 22:23:49 -07:00 |
linmin
|
f43f555dc0
|
add tests for lmdb of datalayer (copied from test_data_layer.cpp)
|
2014-06-12 22:23:49 -07:00 |
linmin
|
12f6fd52a8
|
add option for lmdb
|
2014-06-12 22:23:49 -07:00 |