Evan Shelhamer [Tue, 1 Mar 2016 05:15:56 +0000 (21:15 -0800)]
check all net upgrade conditions
check all conditions all the time; V0 -> V1 and V1 -> V2 do not suffice.
Evan Shelhamer [Tue, 1 Mar 2016 05:13:27 +0000 (21:13 -0800)]
output all logging from upgrade net tools
output info, warnings, and errors for fuller description of the upgrade
Evan Shelhamer [Mon, 29 Feb 2016 17:13:01 +0000 (09:13 -0800)]
Merge pull request #3747 from Nerei/master
[pycaffe] NetSpec: set tops by attribute or dictionary
Anatoly Baksheev [Mon, 29 Feb 2016 15:01:07 +0000 (18:01 +0300)]
NetSpec: allow setting blob names by string
Jeff Donahue [Sun, 28 Feb 2016 08:31:17 +0000 (00:31 -0800)]
Merge pull request #3725 from shaibagon/drop_nd_blobs
supporting N-D Blobs in Dropout layer Reshape
shai [Thu, 25 Feb 2016 10:09:45 +0000 (12:09 +0200)]
supporting N-D Blobs in Dropout layer Reshape
fixing lint errors
Evan Shelhamer [Sat, 27 Feb 2016 20:12:24 +0000 (12:12 -0800)]
Merge pull request #3740 from shelhamer/fix-docker-flags
Fix flags for nvidia-docker from #3518
Evan Shelhamer [Sat, 27 Feb 2016 20:10:16 +0000 (12:10 -0800)]
fix flags in #3518 for nvidia-docker
nvidia-docker requires long args with equal sign as of docker 1.10:
see https://github.com/BVLC/caffe/pull/3518#issuecomment-
189576419
Evan Shelhamer [Sat, 27 Feb 2016 20:08:39 +0000 (12:08 -0800)]
Merge pull request #3518 from zalando/feature/docker_images
[build] Add docker images for running caffe out-of-the-box (caffe:cpu, caffe:gpu)
Evan Lezar [Tue, 5 Jan 2016 15:27:28 +0000 (16:27 +0100)]
Add Dockerfiles for creating Caffe executable images.
These can be used as direct replacements for the Caffe executable.
Evan Shelhamer [Sat, 27 Feb 2016 07:20:39 +0000 (23:20 -0800)]
Merge pull request #3211 from shelhamer/input-layer
Add Input Layer to Replace `input`s
Evan Shelhamer [Sat, 27 Feb 2016 06:20:50 +0000 (22:20 -0800)]
Deprecate ForwardPrefilled(), Forward(bottom, loss) in lieu of dropping
Relax removal of `Forward()` variations by deprecating instead.
Evan Shelhamer [Fri, 4 Dec 2015 03:39:19 +0000 (19:39 -0800)]
[examples] switch examples + models to Input layers
Evan Shelhamer [Thu, 3 Dec 2015 02:31:26 +0000 (18:31 -0800)]
collect Net inputs from Input layers
Restore the list of net inputs for compatibility with the
pycaffe and matcaffe interfaces and downstream C++.
Evan Shelhamer [Sat, 17 Oct 2015 04:11:32 +0000 (21:11 -0700)]
drop Net inputs + Forward with bottoms
Drop special cases for `input` fields, the `Net` input members,
and the `Net` interface for Forward with bottoms along with
Forward() / ForwardPrefilled() distinction.
Evan Shelhamer [Fri, 16 Oct 2015 23:44:45 +0000 (16:44 -0700)]
deprecate input fields and upgrade automagically
Evan Shelhamer [Fri, 16 Oct 2015 23:44:23 +0000 (16:44 -0700)]
add InputLayer for Net input
Create an input layer to replace oddball Net `input` fields.
Jeff Donahue [Thu, 25 Feb 2016 19:17:19 +0000 (11:17 -0800)]
Merge pull request #3612 from kashefy/tied_weights_ip_transpose
Tied weights with transpose flag for InnerProduct layer
Evan Shelhamer [Thu, 25 Feb 2016 07:21:44 +0000 (23:21 -0800)]
Merge pull request #3715 from olesalscheider/master
CMake: Do not include "${PROJECT_BINARY_DIR}/include" with SYSTEM option
Evan Shelhamer [Thu, 25 Feb 2016 01:37:56 +0000 (17:37 -0800)]
Merge pull request #3703 from shaibagon/pycaffe_nd_blobs
[pycaffe] make pycaffe support N-D blobs throughout
Evan Shelhamer [Thu, 25 Feb 2016 00:41:12 +0000 (16:41 -0800)]
Merge pull request #3719 from shelhamer/new-tutorial-notebooks
[example] Improve Tutorial Notebooks on Classification, Learning LeNet, and Fine-tuning
Niels Ole Salscheider [Wed, 24 Feb 2016 16:00:58 +0000 (17:00 +0100)]
CMake: Do not include "${PROJECT_BINARY_DIR}/include" with SYSTEM option
This is important for the include order. Without this patch, a
previously installed caffe.pb.h might be included instead of the one
that is generated during the build.
Jeff Donahue [Wed, 24 Feb 2016 07:42:11 +0000 (23:42 -0800)]
[example] improve brewing logreg notebook
- create solvers inline through python protobuf
- drop manually written solver prototxt
- remove ordering prefix, since there is no real sequencing constraint
for this example
Jeff Donahue [Wed, 24 Feb 2016 07:43:31 +0000 (23:43 -0800)]
[example] improve fine-tuning notebook
- add headings and text detail
- make nets by net spec
- define solvers inline through python protobuf
- do two-stage fine-tuning (first last layer alone, then end-to-end)
- show sample results
Jonathan L Long [Wed, 24 Feb 2016 07:46:49 +0000 (23:46 -0800)]
[example] improve learning LeNet notebook
- add subheadings and list steps for structure
- edit text and comments for clarity
- switch paths and use chdir for idempotency of scripts [shelhamer]
- title accuracy plots, rename ip -> fc to fit common naming, and
rename output layer -> score [shelhamer]
- add experimentation section [shelhamer]
Jonathan L Long [Fri, 5 Feb 2016 03:41:55 +0000 (19:41 -0800)]
[data] get_mnist.sh rewrite; prevents prompt in tutorial notebooks
Jonathan L Long [Wed, 24 Feb 2016 07:34:46 +0000 (23:34 -0800)]
[example] improve classification notebook
- add subheadings and list steps
- edit text, add comments, and try to make the code more understandable
- add new section for summary and encouragement to try your own image
shai [Tue, 23 Feb 2016 08:42:54 +0000 (10:42 +0200)]
removing all references to Blob.num property (that assumes Blob is 4D). Replacing it with accessing Blob.shape[0] - for Blobs with num_axes() != 4
Evan Shelhamer [Sun, 21 Feb 2016 06:42:05 +0000 (22:42 -0800)]
Merge pull request #3696 from shelhamer/fix-osx-install-name
fix library install name on OS X for @rpath linking
Evan Shelhamer [Sun, 21 Feb 2016 06:39:30 +0000 (22:39 -0800)]
Merge pull request #3695 from shelhamer/fix-osx-elcap-cuda-link
Fix OS X El Capitan linking for CUDA
Mohamed Ezz [Fri, 5 Feb 2016 00:54:31 +0000 (01:54 +0100)]
Fix OSX El Capitan CUDA incompatibility, by adding lib to rpath
Evan Shelhamer [Sat, 20 Feb 2016 23:45:31 +0000 (15:45 -0800)]
fix library install name on OSX for relative path linking
for linking of the caffe tools, tests, etc. the library install name
needs to include the @rpath set for the executables and interfaces
this was broken on OSX by #3311
Youssef Kashef [Fri, 29 Jan 2016 18:21:48 +0000 (19:21 +0100)]
tranpose parameter added to IP layer to support tied weights in an autoencoder. Arguments to matrix multiplication function are conditioned on this parameter, no actual transposing takes place.
test ip gradient computation with transpose on
Jeff Donahue [Thu, 18 Feb 2016 07:30:12 +0000 (23:30 -0800)]
Merge pull request #3676 from flx42/leveldb_include_fix
Remove useless LevelDB include
Jeff Donahue [Thu, 18 Feb 2016 04:54:02 +0000 (20:54 -0800)]
Merge pull request #3687 from pra85/patch-2
Fix a typo in docs
Prayag Verma [Thu, 18 Feb 2016 04:43:34 +0000 (10:13 +0530)]
Fix a typo in docs
`peformance` → `performance`
Felix Abecassis [Tue, 16 Feb 2016 00:52:32 +0000 (16:52 -0800)]
Remove useless LevelDB include
The tests could not compile with USE_LEVELDB=0 and LevelDB missing from the system
Evan Shelhamer [Thu, 11 Feb 2016 21:19:50 +0000 (13:19 -0800)]
Merge pull request #3451 from intelcaffe/cmake-clang-fix
Fix to cmake build for clang on linux
Jeff Donahue [Tue, 9 Feb 2016 18:27:35 +0000 (10:27 -0800)]
Merge pull request #3650 from intbots/cmake-matlab-fix
bugfix for incorrect behaviour in `caffe_parse_linker_libs`
Abhijit Kundu [Tue, 9 Feb 2016 07:45:46 +0000 (02:45 -0500)]
bugfix for incorrect behaviour in caffe_parse_linker_libs function while extracting libflags from absolute library path with multiple (dots)
Jeff Donahue [Wed, 3 Feb 2016 06:05:07 +0000 (22:05 -0800)]
Merge pull request #3624 from drnikolaev/bvlc-print-gpu-names
Nicely prints gpu names
Sergei Nikolaev [Tue, 2 Feb 2016 21:48:18 +0000 (13:48 -0800)]
Nicely prints GPU names
Evan Shelhamer [Sat, 30 Jan 2016 03:02:45 +0000 (19:02 -0800)]
Merge pull request #3587 from junshi15/distribute_proto
Include proto definition in distribute dir
Jeff Donahue [Fri, 29 Jan 2016 08:59:23 +0000 (00:59 -0800)]
Merge pull request #3545 from gdh1995/relative-path-for-soft-links
use relative paths on making build/tools/ links
gdh1995 [Wed, 13 Jan 2016 10:20:41 +0000 (18:20 +0800)]
use relative paths on making build/tools/ links
The old uses `abspath`, which I think is so harmful:
* If I `cp -a` the whole project, `build/tools/caffe` still refer to
the old file, until `make clean`, making debugging very hard
* For `tar` and `scp`, the soft links can not work
unless the target project folder has the same path
Jeff Donahue [Wed, 27 Jan 2016 18:43:29 +0000 (10:43 -0800)]
Merge pull request #2917 from ajkl/patch-2
fixing the database param
Jeff Donahue [Wed, 27 Jan 2016 18:37:48 +0000 (10:37 -0800)]
Merge pull request #3022 from jeffdonahue/expose-param-display-names
Net: expose param_display_names_
Jeff Donahue [Wed, 27 Jan 2016 18:34:20 +0000 (10:34 -0800)]
Merge pull request #3605 from jeffdonahue/bias-gemm-type-fix
Remove incorrect cast of gemm int arg to Dtype in BiasLayer
Jeff Donahue [Wed, 27 Jan 2016 07:09:27 +0000 (23:09 -0800)]
Remove incorrect cast of gemm int arg to Dtype in BiasLayer
Evan Shelhamer [Wed, 27 Jan 2016 01:43:14 +0000 (17:43 -0800)]
Merge pull request #3591 from jeffdonahue/scale-bias-layer
Scale and Bias Layers
Jeff Donahue [Tue, 26 Jan 2016 23:34:42 +0000 (15:34 -0800)]
Merge pull request #2651 from keir/master
Keir Mierle [Fri, 26 Jun 2015 07:10:21 +0000 (00:10 -0700)]
Make the two separate build systems clearer in the documentation
Jeff Donahue [Tue, 26 Jan 2016 23:20:07 +0000 (15:20 -0800)]
Merge pull request #3602 from jeffdonahue/rm-cuda-props
Remove unnecessary CAFFE_TEST_CUDA_PROP declarations
Jeff Donahue [Tue, 26 Jan 2016 23:17:00 +0000 (15:17 -0800)]
Merge pull request #2144 from tishibas/load_image-improved
improved to load RGB image as grayscale image
Jeff Donahue [Tue, 26 Jan 2016 23:07:43 +0000 (15:07 -0800)]
Merge pull request #2810 from madan-ram/patch-2
Madan Ram [Thu, 23 Jul 2015 11:12:15 +0000 (16:42 +0530)]
Update mnist readme.md: scale moved to transform_param
Jeff Donahue [Tue, 26 Jan 2016 22:32:39 +0000 (14:32 -0800)]
Merge pull request #3297 from sjbrown/patch-1
Add a -c to wget so that it continues interrupted downloads
Jeff Donahue [Tue, 26 Jan 2016 22:27:27 +0000 (14:27 -0800)]
Merge pull request #3132 from bwilbertz/LastIterationLoss
Fix loss of last iteration when average_loss > 1
Jeff Donahue [Tue, 26 Jan 2016 21:58:58 +0000 (13:58 -0800)]
Remove unnecessary CAFFE_TEST_CUDA_PROP declarations
Jeff Donahue [Tue, 26 Jan 2016 18:01:34 +0000 (10:01 -0800)]
Merge pull request #3393 from kkhoot/fix-reshape
Kang Kim [Fri, 27 Nov 2015 12:57:51 +0000 (21:57 +0900)]
Prevent in-place computation in ReshapeLayer and FlattenLayer
Jeff Donahue [Tue, 26 Jan 2016 17:55:08 +0000 (09:55 -0800)]
Merge pull request #3496 from jeffdonahue/fix-testdatatransformer-leaks
TestDataTransformer: fix some memory leaks
Jeff Donahue [Tue, 26 Jan 2016 17:38:20 +0000 (09:38 -0800)]
Merge pull request #3581 from Austriker/draw-python3
Updated draw.py to make it work under Python 3
Hugo Serrat [Thu, 21 Jan 2016 13:34:01 +0000 (14:34 +0100)]
Updated import to make it work with pydotplus
Ronghang Hu [Sat, 23 Jan 2016 20:15:46 +0000 (12:15 -0800)]
Merge pull request #3593 from ronghanghu/matcaffe-versioning
show Caffe's version from MatCaffe
Ronghang Hu [Sat, 23 Jan 2016 09:22:04 +0000 (01:22 -0800)]
show Caffe's version from MatCaffe
Jeff Donahue [Fri, 22 Jan 2016 23:57:47 +0000 (15:57 -0800)]
Separation and generalization of ChannelwiseAffineLayer into BiasLayer
and ScaleLayer. The behavior of ChannelwiseAffineLayer can be
reproduced by a ScaleLayer with `scale_param { bias_term: true }`.
BiasLayer and ScaleLayer each take 1 or 2 bottoms, with the output having
the same shape as the first. The second input -- either another bottom or a
learned parameter -- will have its axes (virtually) broadcast and tiled to have
the same shape as the first, after which elementwise addition (Bias) or
multiplication (Scale) is performed.
Evan Shelhamer [Sat, 23 Jan 2016 02:54:17 +0000 (18:54 -0800)]
Merge pull request #3311 from lukeyeager/bvlc/versioning
Add versioning to build and mark version 1.0.0-rc3
Philipp Krähenbühl [Sat, 23 Jan 2016 00:59:01 +0000 (16:59 -0800)]
Merge pull request #3584 from graphaelli/python3-config
Add makefile config option for linking Python 3 libraries
Luke Yeager [Fri, 22 Jan 2016 23:30:35 +0000 (15:30 -0800)]
Version 1.0.0-rc3
Dmytro Mishkin [Thu, 14 Jan 2016 15:10:11 +0000 (17:10 +0200)]
Add ChannelwiseAffine for batch norm
Jun Shi [Fri, 22 Jan 2016 13:09:21 +0000 (05:09 -0800)]
copy proto to distribute directory
Evan Shelhamer [Fri, 22 Jan 2016 18:49:48 +0000 (10:49 -0800)]
Merge pull request #3388 from mohomran/exponential_linear_units
Exponential Linear Units
Robbie Cooper [Thu, 21 Jan 2016 19:11:00 +0000 (14:11 -0500)]
Add makefile config option for linking Python 3 libraries
Jon Long [Wed, 20 Jan 2016 21:38:10 +0000 (13:38 -0800)]
Merge pull request #3536 from intelcaffe/im2col-speedup
Performance related update of im2col() and col2im() functions
Evan Shelhamer [Wed, 20 Jan 2016 18:09:59 +0000 (10:09 -0800)]
Merge pull request #3574 from tahtguymike/maxpool_bug
Substitute Caffe engine for cuDNN max pooling
as workaround for in-place after max pooling issue
thatguymike [Wed, 20 Jan 2016 01:01:34 +0000 (17:01 -0800)]
Workaround for inplace max pooling issue
Mariusz Moczala [Wed, 20 Jan 2016 08:28:35 +0000 (09:28 +0100)]
Performance related update of im2col() and col2im() functions
Jon Long [Fri, 8 Jan 2016 02:48:55 +0000 (18:48 -0800)]
Merge pull request #3525 from philkr/cmake_python3
CMake python version fix
Jon Long [Fri, 8 Jan 2016 02:42:40 +0000 (18:42 -0800)]
Merge pull request #2865 from philkr/top_bottom_names
Exposing layer top and bottom names to python
philkr [Wed, 6 Jan 2016 15:23:35 +0000 (07:23 -0800)]
CMake python version fix
Jeff Donahue [Wed, 6 Jan 2016 05:20:54 +0000 (21:20 -0800)]
Merge pull request #3519 from philkr/faster_solver
Speeding up ADAM (and maybe other solvers too)
philkr [Tue, 5 Jan 2016 20:45:52 +0000 (12:45 -0800)]
Speeding up the GPU solvers
philkr [Wed, 5 Aug 2015 18:54:08 +0000 (11:54 -0700)]
Exposing layer top and bottom names to python
Jeff Donahue [Wed, 30 Dec 2015 05:10:14 +0000 (21:10 -0800)]
TestDataTransformer: fix some memory leaks caused by use of 'new'
Jeff Donahue [Tue, 29 Dec 2015 06:27:34 +0000 (22:27 -0800)]
Merge pull request #3490 from fyu/spaceplus
remove extra space before +
Fisher Yu [Tue, 29 Dec 2015 03:46:49 +0000 (22:46 -0500)]
remove extra space before +
Jon Long [Mon, 28 Dec 2015 23:56:46 +0000 (15:56 -0800)]
Merge pull request #3487 from longjon/dilation
An early implementation of this functionality for Caffe was written by @gpapan, which was extended and improved by @tamakoji in a previous implementation of this branch.
Jonathan L Long [Mon, 28 Dec 2015 04:56:24 +0000 (20:56 -0800)]
enable dilated deconvolution
Since the underlying routines are shared, we need only upgrade
compute_output_shape.
Jonathan L Long [Sat, 26 Dec 2015 21:10:11 +0000 (13:10 -0800)]
add short description of dilation to caffe.proto
Fisher Yu [Sat, 26 Dec 2015 21:10:02 +0000 (13:10 -0800)]
disable dilated deconvolution
Fisher Yu [Sat, 26 Dec 2015 21:04:25 +0000 (13:04 -0800)]
add and improve tests for dilated convolution/im2col
Fisher Yu [Mon, 28 Dec 2015 04:48:30 +0000 (20:48 -0800)]
add support for N-D dilated convolution
Fisher Yu [Mon, 7 Dec 2015 01:04:43 +0000 (20:04 -0500)]
add support for 2D dilated convolution
Jon Long [Sun, 20 Dec 2015 23:49:58 +0000 (15:49 -0800)]
Merge pull request #3468 from mnogu/blobs_lr-to-lr_mult
Replace blobs_lr with lr_mult in readme.md.
Muneyuki Noguchi [Sun, 20 Dec 2015 10:12:09 +0000 (19:12 +0900)]
Replace blobs_lr with lr_mult in readme.md.
models/finetune_flickr_style/deploy.prototxt uses lr_mult now.
Jacek Czaja [Mon, 14 Dec 2015 15:45:59 +0000 (16:45 +0100)]
- Fix to cmake build for clang
Evan Shelhamer [Sat, 12 Dec 2015 04:00:00 +0000 (01:00 -0300)]
Merge pull request #3439 from flx42/cudnn_v4
Add ifdef in CuDNNConvolutionLayer for cuDNN v4
Felix Abecassis [Thu, 10 Dec 2015 23:11:51 +0000 (15:11 -0800)]
Fix CuDNNConvolutionLayer for cuDNN v4
Add a macro to check the current cuDNN version
Evan Shelhamer [Wed, 9 Dec 2015 20:31:45 +0000 (12:31 -0800)]
Merge pull request #3432 from ianfhunter/patch-1
[docs] fix typo in interfaces.md