Jon Long [Wed, 4 Mar 2015 11:33:13 +0000 (03:33 -0800)]
Merge pull request #1966 from philkr/python_fixes
cmake and python3 bugfixes for #1939 and #1923
Evan Shelhamer [Wed, 4 Mar 2015 06:27:55 +0000 (22:27 -0800)]
Merge pull request #1970 from jeffdonahue/tensor-blob
Blobs are N-D arrays (for N not necessarily equals 4)
Jonathan L Long [Mon, 2 Mar 2015 23:54:11 +0000 (15:54 -0800)]
[pytest] use non-4d blobs in test_python_layer
Jonathan L Long [Mon, 2 Mar 2015 23:27:45 +0000 (15:27 -0800)]
[pycaffe] expose Blob.reshape as *args function
Jeff Donahue [Sat, 31 Jan 2015 07:16:44 +0000 (23:16 -0800)]
Add option not to reshape to Blob::FromProto; use when loading Blobs
from saved NetParameter
Want to keep the param Blob shape the layer has set, and not necessarily
adopt the one from the saved net (e.g. want to keep new 1D bias shape,
rather than take the (1 x 1 x 1 x D) shape from a legacy net).
Jeff Donahue [Fri, 2 Jan 2015 01:32:38 +0000 (17:32 -0800)]
PyBlobs support generalized axes
Jeff Donahue [Fri, 16 Jan 2015 03:50:42 +0000 (19:50 -0800)]
Add CHECK_EQ(4, ...)s to "vision layers" to enforce that the
num/channnels/height/width indexing is valid.
Jeff Donahue [Thu, 1 Jan 2015 00:06:46 +0000 (16:06 -0800)]
DummyDataLayer outputs blobs of arbitrary shape
Jeff Donahue [Sun, 30 Nov 2014 02:00:44 +0000 (18:00 -0800)]
EuclideanLossLayer: generalized Blob axes
Jeff Donahue [Wed, 26 Nov 2014 20:57:15 +0000 (12:57 -0800)]
WindowDataLayer outputs 1D labels
Jeff Donahue [Wed, 26 Nov 2014 20:56:14 +0000 (12:56 -0800)]
ImageDataLayer outputs 1D labels
Jeff Donahue [Wed, 26 Nov 2014 13:42:50 +0000 (05:42 -0800)]
MemoryDataLayer outputs 1D labels
Jeff Donahue [Wed, 26 Nov 2014 13:34:47 +0000 (05:34 -0800)]
DataLayer outputs 1D labels
Jeff Donahue [Wed, 26 Nov 2014 13:42:11 +0000 (05:42 -0800)]
HDF5DataLayer shapes output according to HDF5 shape
Jeff Donahue [Wed, 26 Nov 2014 13:02:15 +0000 (05:02 -0800)]
SplitLayer: change Reshape(n,h,c,w) to ReshapeLike(...)
Jeff Donahue [Sat, 31 Jan 2015 07:22:26 +0000 (23:22 -0800)]
SoftmaxLossLayer generalized like SoftmaxLayer
Jeff Donahue [Tue, 10 Feb 2015 02:12:54 +0000 (18:12 -0800)]
CuDNNSoftmaxLayer: generalized Blob axes
Jeff Donahue [Sun, 15 Feb 2015 21:26:36 +0000 (13:26 -0800)]
SoftmaxLayer: generalized Blob axes
Jeff Donahue [Wed, 26 Nov 2014 11:22:59 +0000 (03:22 -0800)]
SliceLayer: generalized Blob axes
Jeff Donahue [Wed, 26 Nov 2014 08:03:36 +0000 (00:03 -0800)]
ConcatLayer: generalized Blob axes
Jeff Donahue [Wed, 26 Nov 2014 10:24:41 +0000 (02:24 -0800)]
TestConcatLayer: add forward/gradient tests for concatenation along num
Jeff Donahue [Wed, 26 Nov 2014 10:12:09 +0000 (02:12 -0800)]
TestConcatLayer: fix style errors
Jeff Donahue [Wed, 26 Nov 2014 11:23:42 +0000 (03:23 -0800)]
common_layers.hpp: remove unused "Blob col_bob_"
Jeff Donahue [Wed, 26 Nov 2014 08:11:06 +0000 (00:11 -0800)]
FlattenLayer: generalized Blob axes
Jeff Donahue [Wed, 26 Nov 2014 07:47:10 +0000 (23:47 -0800)]
EltwiseLayer need not assume old 4D dim names
Jeff Donahue [Fri, 13 Feb 2015 23:52:39 +0000 (15:52 -0800)]
Test{Net,Solver} fixes for AccuracyLayer generalization
Jeff Donahue [Sat, 31 Jan 2015 07:23:27 +0000 (23:23 -0800)]
AccuracyLayer generalized to N instance axes
Jeff Donahue [Sun, 30 Nov 2014 00:48:32 +0000 (16:48 -0800)]
AccuracyLayer output is 0D (scalar)
Jeff Donahue [Wed, 26 Nov 2014 07:46:57 +0000 (23:46 -0800)]
LossLayer output is 0D (scalar)
Jeff Donahue [Sun, 30 Nov 2014 01:02:18 +0000 (17:02 -0800)]
ConvLayer biases are 1D
Jeff Donahue [Sat, 29 Nov 2014 22:26:48 +0000 (14:26 -0800)]
InnerProductLayer can multiply along any axis
Jeff Donahue [Mon, 16 Feb 2015 09:29:17 +0000 (01:29 -0800)]
Fix sparse GaussianFiller for new IPLayer weight axes
Jeff Donahue [Wed, 26 Nov 2014 06:48:03 +0000 (22:48 -0800)]
InnerProductLayer weights are 2D; biases are 1D
Jeff Donahue [Sun, 30 Nov 2014 00:11:07 +0000 (16:11 -0800)]
TestBlob: test that legacy BlobProtos are correctly handled by ShapeEquals
Jeff Donahue [Wed, 4 Feb 2015 23:00:03 +0000 (15:00 -0800)]
add offset, {data,diff}_at nd blob accessors
Jeff Donahue [Thu, 1 Jan 2015 00:05:52 +0000 (16:05 -0800)]
Add BlobShape message; use for Net input shapes
Jeff Donahue [Wed, 26 Nov 2014 02:17:45 +0000 (18:17 -0800)]
Blobs are ND arrays (for N not necessarily equals 4).
vector<int> shape_ instead of (num, channels, height, width).
Jon Long [Fri, 27 Feb 2015 18:32:49 +0000 (10:32 -0800)]
Merge pull request #1999 from boechat107/patch-2
Small fix (visualization) on SLICE layer's documentation
Evan Shelhamer [Thu, 26 Feb 2015 00:37:55 +0000 (16:37 -0800)]
Merge pull request #1955 from philkr/reshaping_encoded
Decoding the datum before feeding it into the reshaping data layer
Andre Ambrosio Boechat [Wed, 25 Feb 2015 17:56:18 +0000 (14:56 -0300)]
Small fix (visualization) on SLICE layer's documentation
The sample was missing some additional spaces to be correctly rendered on the HTML. The mistake was mine.
philkr [Wed, 25 Feb 2015 16:17:06 +0000 (08:17 -0800)]
Fixing two bugs related to python3 and PROJECT_SOURCE_DIR
Jon Long [Wed, 25 Feb 2015 07:00:27 +0000 (23:00 -0800)]
Merge pull request #1960 from gustavla/makefile_fix
Replaced illegal tab in Makefile with spaces.
Evan Shelhamer [Wed, 25 Feb 2015 06:05:06 +0000 (22:05 -0800)]
Merge pull request #1961 from sergeyk/master
fix accelerate / veclib path for OS X 10.10
Sergey Karayev [Wed, 25 Feb 2015 05:52:32 +0000 (21:52 -0800)]
Makefile fix for OS X 10.10
Gustav Larsson [Wed, 25 Feb 2015 05:22:15 +0000 (23:22 -0600)]
Replaced illegal tab in Makefile with spaces.
Commands, such as $(error ...), are not allowed to be indented with tabs
outside of targets, throwing an error instead of outputting the actual
error. The solution is to use innocuous spaces instead. Ideally, spaces
should be used everywhere outside targets, but since make does not mind
it if variable assignments are tab-indented outside targets, a complete
overhaul is not necessary. However, if more errors are added, it might
make more sense to be consistent.
Also, make will already add a period so I removed it.
forresti [Wed, 25 Feb 2015 00:51:56 +0000 (16:51 -0800)]
fixed matcaffe printout to specify num of args (now including train/test phase)
philkr [Wed, 25 Feb 2015 00:02:06 +0000 (16:02 -0800)]
Decoding the datum before feeding it into the reshaping data layer
Jon Long [Tue, 24 Feb 2015 23:00:58 +0000 (15:00 -0800)]
Merge pull request #1923 from philkr/python3_master
Making python3 work with cmake and the new python wrapper
philkr [Tue, 17 Feb 2015 18:50:12 +0000 (10:50 -0800)]
Making python3 work with cmake and the new python wrapper
Jon Long [Tue, 24 Feb 2015 04:04:48 +0000 (20:04 -0800)]
Merge pull request #1926 from shelhamer/test-caffe-tool
Check caffe tool runs
Jon Long [Tue, 24 Feb 2015 04:01:27 +0000 (20:01 -0800)]
Merge pull request #1941 from jsupancic/cpp_lint_python2
fixes: cpp_lint.py fails silently with Python3
Jon Long [Tue, 24 Feb 2015 03:52:12 +0000 (19:52 -0800)]
Merge pull request #1948 from spmallick/patch-1
APPLE was misspelled in Line 27
spmallick [Tue, 24 Feb 2015 00:13:20 +0000 (16:13 -0800)]
APPLE was misspelled. in Line 27
Evan Shelhamer [Sun, 22 Feb 2015 18:35:10 +0000 (10:35 -0800)]
Merge pull request #1939 from Nerei/bugfix/install_rpath_for_pycaffe
[cmake] fix install rpath for pycaffe
James Supancic III [Sun, 22 Feb 2015 18:16:45 +0000 (10:16 -0800)]
cpp_lint.py fails silently with Python3 (which is the default on some systems).
This commit specifies Python2 with which cpp_lint.py works :-)
Anatoly Baksheev [Sun, 22 Feb 2015 17:14:40 +0000 (20:14 +0300)]
minor cmake sumamry log fix
Anatoly Baksheev [Sun, 22 Feb 2015 16:04:22 +0000 (19:04 +0300)]
fixed bug in install-tree: _caffe.so installed by install(TARGET ...) was overwritten with symlink created at build time and installed with install(DIRECTORY ...)
Anatoly Baksheev [Sun, 22 Feb 2015 16:03:47 +0000 (19:03 +0300)]
set proper CMAKE_INSTALL_RPATH for _caffe.so and tools
Anatoly Baksheev [Sun, 22 Feb 2015 15:58:12 +0000 (18:58 +0300)]
ignore pycharm files
Evan Shelhamer [Sat, 21 Feb 2015 04:29:53 +0000 (20:29 -0800)]
check caffe tool runs in runtest
Evan Shelhamer [Fri, 20 Feb 2015 20:20:23 +0000 (12:20 -0800)]
Merge pull request #1921 from shelhamer/fix-tool-linking
[build] fix dynamic linking of tools
Evan Shelhamer [Fri, 20 Feb 2015 19:18:47 +0000 (11:18 -0800)]
[build] fix dynamic linking of tools
set the right rpath for tools and examples respectively
thanks for the report @mees!
Evan Shelhamer [Fri, 20 Feb 2015 06:16:10 +0000 (22:16 -0800)]
Merge pull request #1914 from eerwitt/master
fix Imagenet example path
e3 [Fri, 20 Feb 2015 06:07:27 +0000 (22:07 -0800)]
Updated the path for get_ilsvrc_aux.sh to match what is found in the current project
Evan Shelhamer [Fri, 20 Feb 2015 02:35:23 +0000 (18:35 -0800)]
Merge pull request #1849 from BVLC/next
Next: release candidater
Brandon Amos [Mon, 16 Feb 2015 20:09:24 +0000 (15:09 -0500)]
Correct 'epochs' to 'iterations'
See https://github.com/BVLC/caffe/blob/master/models/bvlc_reference_caffenet/solver.prototxt
Andre Ambrosio Boechat [Fri, 30 Jan 2015 15:57:31 +0000 (13:57 -0200)]
Brief explanation of SLICE layer's attributes
* A sample code was added.
* `slice_dim` and `slice_point` attributes were explained.
Evan Shelhamer [Thu, 29 Jan 2015 18:53:44 +0000 (10:53 -0800)]
[docs] add check mode hint to CPU-only mode error
Evan Shelhamer [Thu, 29 Jan 2015 17:26:12 +0000 (09:26 -0800)]
[docs] send API link to class list
Evan Shelhamer [Fri, 20 Feb 2015 02:14:57 +0000 (18:14 -0800)]
[build] fix rpath for examples
Evan Shelhamer [Fri, 20 Feb 2015 02:01:13 +0000 (18:01 -0800)]
Merge pull request #1910 from philkr/encoded
add force_encoded_color flag to the data layer and warn about
mixed encoding
Evan Shelhamer [Fri, 20 Feb 2015 01:54:49 +0000 (17:54 -0800)]
Repeal revert of #1878
Evan Shelhamer [Fri, 20 Feb 2015 01:49:12 +0000 (17:49 -0800)]
[docs] add gitter chat badge
philkr [Thu, 19 Feb 2015 21:46:21 +0000 (13:46 -0800)]
added a force_encoded_color flag to the data layer. Printing a warning if images of different channel dimensions are encoded together
Evan Shelhamer [Thu, 19 Feb 2015 19:03:23 +0000 (11:03 -0800)]
Revert "Merge pull request #1878 from philkr/encoded"
This reverts the encoding cleanup since it breaks data processing for
existing inputs as discussed in #1901.
Evan Shelhamer [Thu, 19 Feb 2015 18:56:26 +0000 (10:56 -0800)]
Merge pull request #1907 from shelhamer/memory-warn-transform
relax MemoryData transform check to warning
Evan Shelhamer [Thu, 19 Feb 2015 18:31:23 +0000 (10:31 -0800)]
relax MemoryData transform check to warning
warning is still useful to keep from accidentally running data without
pre-processing.
Evan Shelhamer [Thu, 19 Feb 2015 08:37:53 +0000 (00:37 -0800)]
Merge pull request #1899 from philkr/project_source_dir
[cmake] CMAKE_SOURCE/BINARY_DIR to PROJECT_SOURCE/BINARY_DIR
Evan Shelhamer [Thu, 19 Feb 2015 08:25:18 +0000 (00:25 -0800)]
Merge pull request #1902 from shelhamer/py-stopgap
Stop-gap fixes for pycaffe
Evan Shelhamer [Thu, 19 Feb 2015 04:53:22 +0000 (20:53 -0800)]
[pycaffe] switch examples to Transformer
This is a quick translation of the examples to the caffe.io.Transformer
interface. The results are not strictly identical to the earlier
implementation! The models now use a mean pixel instead of a mean image
for simplicity. The output classifications and detections are preserved
but scores may differ.
Note: the examples will be rewritten to make use of `caffe.Net` alone
since it is the true interface, but not yet.
Evan Shelhamer [Thu, 19 Feb 2015 04:47:12 +0000 (20:47 -0800)]
[pycaffe] take pre-processing from Net and give to Transformer
The pyreformation #1703 aligned pycaffe Net with the real Caffe Net.
This broke the pre-processing helpers that had been bolted on to
`caffe.Net`.
These responsibilities are gathered in `caffe.io.Transformer` instead.
This is only an intermediate step in a real solution to pycaffe IO that
should make use of the same data pipeline as the rest of the framework.
See #1245 for further thoughts on input processing.
Evan Shelhamer [Thu, 19 Feb 2015 04:46:41 +0000 (20:46 -0800)]
[pycaffe] import newline cleanup
philkr [Wed, 18 Feb 2015 16:38:33 +0000 (08:38 -0800)]
Changing CMAKE_SOURCE/BINARY_DIR to PROJECT_SOURCE/BINARY_DIR
Jonathan L Long [Wed, 18 Feb 2015 02:23:32 +0000 (18:23 -0800)]
comment fix: Decaf -> Caffe
Jonathan L Long [Tue, 17 Feb 2015 23:46:29 +0000 (15:46 -0800)]
[pycaffe] fix bug in Python layer setup
Previously, setup did not correctly override LayerSetUp (and therefore
did not run) due to an incorrect type.
Evan Shelhamer [Tue, 17 Feb 2015 20:02:55 +0000 (12:02 -0800)]
Merge pull request #1790 from shelhamer/net-phase
Give phase to Net
Evan Shelhamer [Mon, 26 Jan 2015 06:06:23 +0000 (22:06 -0800)]
[matcaffe] give phase to Net
Evan Shelhamer [Mon, 26 Jan 2015 04:57:23 +0000 (20:57 -0800)]
[pycaffe] give phase to Net
- expose `caffe.{TRAIN,TEST}` constants
- instantiate `caffe.Net`s with phase
- drop singleton phase interface `caffe.set_phase_{train,test}`
Evan Shelhamer [Sat, 24 Jan 2015 06:49:38 +0000 (22:49 -0800)]
tools make net with phase
Evan Shelhamer [Sat, 24 Jan 2015 06:48:42 +0000 (22:48 -0800)]
construct Net from file and phase
Evan Shelhamer [Fri, 23 Jan 2015 09:33:09 +0000 (01:33 -0800)]
no phase for the solver to orchestrate
Evan Shelhamer [Fri, 23 Jan 2015 09:31:53 +0000 (01:31 -0800)]
pass phase to transformer through layer
Evan Shelhamer [Fri, 23 Jan 2015 08:31:28 +0000 (00:31 -0800)]
give phase to Net and Layer
Give the responsibility for phase to Net and Layer, making phase an
immutable choice at instantiation and dropping it from the Caffe
singleton.
Evan Shelhamer [Tue, 17 Feb 2015 09:34:31 +0000 (01:34 -0800)]
Merge pull request #1703 from longjon/pyreformation
Reform the boost::python wrapper, including layers implemented in Python
Evan Shelhamer [Tue, 17 Feb 2015 09:20:42 +0000 (01:20 -0800)]
Merge pull request #1887 from shelhamer/docs
Documentation: README, credits, favicon, installation, and workflow
Jonathan L Long [Tue, 17 Feb 2015 09:19:51 +0000 (01:19 -0800)]
[make] link libcaffe.so before dependencies
Evan Shelhamer [Tue, 17 Feb 2015 09:07:05 +0000 (01:07 -0800)]
[docs] explain one true branch and dev workflow
Evan Shelhamer [Tue, 17 Feb 2015 08:46:14 +0000 (00:46 -0800)]
[docs] update and split installation
Evan Shelhamer [Tue, 10 Feb 2015 21:26:11 +0000 (13:26 -0800)]
switch to V2 proto definitions for pytest
Jonathan L Long [Sat, 10 Jan 2015 09:51:39 +0000 (01:51 -0800)]
[travis] enable Python layer for testing