Carl Doersch [Sun, 3 Jul 2016 19:32:19 +0000 (12:32 -0700)]
Add phase support for draw net
Jeff Donahue [Fri, 17 Jun 2016 23:35:07 +0000 (16:35 -0700)]
Merge pull request #3020 from philkr/python_solver_callback
Exposing solver callbacks to python
philkr [Thu, 3 Sep 2015 21:28:55 +0000 (14:28 -0700)]
Exposing solver callbacks to python
Evan Shelhamer [Fri, 3 Jun 2016 01:29:36 +0000 (18:29 -0700)]
Merge pull request #4227 from philkr/save_hdf5
[pycaffe] expose saving/loading nets as hdf5 to python
philkr [Tue, 24 May 2016 03:09:45 +0000 (20:09 -0700)]
Exposing load_hdf5 and save_hdf5 to python
Evan Shelhamer [Thu, 2 Jun 2016 19:28:51 +0000 (12:28 -0700)]
Merge pull request #4259 from chuckcho/fixing-typo
Fixing a typo
Chuck Cho [Thu, 2 Jun 2016 18:35:14 +0000 (14:35 -0400)]
Fixing a typo
Evan Shelhamer [Thu, 2 Jun 2016 00:35:13 +0000 (17:35 -0700)]
Merge pull request #4254 from lukeyeager/bvlc/travis-caching
[build] Cache protobuf3 build in TravisCI
Jeff Donahue [Thu, 2 Jun 2016 00:28:25 +0000 (17:28 -0700)]
Merge pull request #3948 from jeffdonahue/recurrent-layer
RNN + LSTM Layers
Jeff Donahue [Tue, 5 Apr 2016 16:56:04 +0000 (09:56 -0700)]
Add LSTMLayer and LSTMUnitLayer, with tests
Jeff Donahue [Sun, 15 Feb 2015 22:56:50 +0000 (14:56 -0800)]
Add RNNLayer, with tests
Jeff Donahue [Sun, 15 Feb 2015 22:28:01 +0000 (14:28 -0800)]
Add RecurrentLayer: an abstract superclass for other recurrent layer types
Luke Yeager [Wed, 1 Jun 2016 17:37:14 +0000 (10:37 -0700)]
Cache protobuf3 build in TravisCI
Evan Shelhamer [Tue, 31 May 2016 00:04:16 +0000 (17:04 -0700)]
Merge pull request #4236 from CDLuminate/fix-spelling-error
fix spelling error in memory_data_layer.cpp
Jon Long [Mon, 30 May 2016 21:10:05 +0000 (14:10 -0700)]
Merge pull request #4218 from malreddysid/master
Resolve SIGSEGV error in image_data_layer.cpp
Lumin Zhou [Mon, 30 May 2016 04:14:42 +0000 (04:14 +0000)]
fix spelling error in memory_data_layer.cpp
Siddarth Malreddy [Thu, 26 May 2016 18:01:31 +0000 (23:31 +0530)]
Check for non-empty ImageData filelist.
Evan Shelhamer [Thu, 26 May 2016 01:11:15 +0000 (18:11 -0700)]
Merge pull request #4214 from lukeyeager/bvlc/remove-comment-in-tests
Remove misleading comment from a test file
Luke Yeager [Wed, 25 May 2016 23:39:55 +0000 (16:39 -0700)]
Remove misleading comment from a test file
Evan Shelhamer [Wed, 25 May 2016 21:45:19 +0000 (14:45 -0700)]
Merge pull request #4207 from lukeyeager/bvlc/travis-overhaul
[testing] Overhaul TravisCI
Luke Yeager [Tue, 24 May 2016 17:36:23 +0000 (10:36 -0700)]
Overhaul TravisCI
* Run on Ubuntu 14.04
* Test cuDNN builds
* Build with OpenBLAS
NOTE: Python3 build only works with CMake
Jon Long [Wed, 18 May 2016 18:36:18 +0000 (11:36 -0700)]
Merge pull request #4059 from crazytan/master
handle spaces in image file names
Thanks @bchu for an earlier implementation.
crazytan [Wed, 27 Apr 2016 05:01:30 +0000 (01:01 -0400)]
handle image names with spaces
Jon Long [Tue, 17 May 2016 22:51:52 +0000 (15:51 -0700)]
Merge pull request #4070 from crazytan/ipython
fix problems in net_surgery.ipynb
Jon Long [Tue, 17 May 2016 22:33:21 +0000 (15:33 -0700)]
Merge pull request #4146 from yalesong/fix-makefile-osx-yosemite
Fix Makefile CUDA_VERSION extraction on OSX Yosemite
Jon Long [Tue, 17 May 2016 22:07:42 +0000 (15:07 -0700)]
Merge pull request #4148 from bobpoekert/window_data_nonzero_check
Window data foreground/background nonzero check
Evan Shelhamer [Tue, 17 May 2016 19:14:39 +0000 (12:14 -0700)]
Merge pull request #4159 from flx42/cudnn_v5_support
Add cuDNN v5 support, drop cuDNN v3 support
Felix Abecassis [Mon, 16 May 2016 21:35:40 +0000 (14:35 -0700)]
Update supported cuDNN version in the documentation
Felix Abecassis [Mon, 16 May 2016 21:32:34 +0000 (14:32 -0700)]
Update Dockerfile to cuDNN v5
Felix Abecassis [Mon, 16 May 2016 21:03:38 +0000 (14:03 -0700)]
Add cuDNN v5 support, drop cuDNN v3 support
cuDNN v4 is still supported.
Bob Poekert [Sat, 14 May 2016 05:06:33 +0000 (22:06 -0700)]
add check for background and foreground window size > 0 in WindowData layer
Evan Shelhamer [Fri, 13 May 2016 20:16:25 +0000 (13:16 -0700)]
Merge pull request #4144 from millskyle/python_io_typo
[bug] typo in io.py
Yale Song [Fri, 13 May 2016 20:06:59 +0000 (16:06 -0400)]
Fix Makefile CUDA_VERSION extraction on OSX Yosemite
Kyle Mills [Fri, 13 May 2016 15:15:33 +0000 (11:15 -0400)]
fixed typo in io.py
Jon Long [Thu, 12 May 2016 06:04:22 +0000 (23:04 -0700)]
Merge pull request #4128 from gdh1995/master
a comment misses a space char
gdh1995 [Wed, 11 May 2016 12:51:07 +0000 (20:51 +0800)]
a comment misses a space char
Evan Shelhamer [Tue, 10 May 2016 23:55:31 +0000 (16:55 -0700)]
Merge pull request #4121 from rayglover-ibm/cmake
[build] CMake customisable version/soversion
ray glover [Tue, 10 May 2016 14:44:47 +0000 (15:44 +0100)]
[build] (CMake) customisable Caffe version/soversion
Evan Shelhamer [Mon, 9 May 2016 21:26:52 +0000 (14:26 -0700)]
Merge pull request #4117 from lukeyeager/bvlc/fix-lmdb-pr
[bug] Catch MDB_MAP_FULL errors from mdb_txn_commit
Luke Yeager [Mon, 9 May 2016 18:21:26 +0000 (11:21 -0700)]
Catch MDB_MAP_FULL errors from mdb_txn_commit
Jon Long [Mon, 9 May 2016 01:13:11 +0000 (18:13 -0700)]
Merge pull request #2079 from longjon/parameter-layer
Add parameter layer for learning any bottom
Jeff Donahue [Fri, 6 May 2016 03:00:34 +0000 (20:00 -0700)]
Merge pull request #4101 from erictzeng/reshape_zero
Allow reshaping blobs to size 0.
Eric Tzeng [Fri, 6 May 2016 01:29:30 +0000 (18:29 -0700)]
Allow reshaping blobs to size 0.
Also add a test that reshapes a blob to shape (0, 5).
crazytan [Thu, 28 Apr 2016 22:45:13 +0000 (18:45 -0400)]
fix problems in net_surgery.ipynb
Jonathan L Long [Thu, 9 Jul 2015 22:49:48 +0000 (15:49 -0700)]
add parameter layer for learning any bottom
Evan Shelhamer [Wed, 4 May 2016 20:47:05 +0000 (13:47 -0700)]
Merge pull request #4094 from shelhamer/make-clean-clears-distribute
Note that `make clean` clears build and distribute dirs
Evan Shelhamer [Wed, 4 May 2016 20:31:35 +0000 (13:31 -0700)]
[build] note that `make clean` clears build and distribute dirs
Jon Long [Wed, 4 May 2016 20:15:24 +0000 (13:15 -0700)]
Merge pull request #4048 from achalddave/python_plot_exit_properly
(Minor) Exit on error and report argument error details
Jon Long [Wed, 4 May 2016 20:06:03 +0000 (13:06 -0700)]
Merge pull request #3995 from ZhouYzzz/python-phase
Allow the python layer have attribute "phase"
Evan Shelhamer [Wed, 4 May 2016 17:19:46 +0000 (10:19 -0700)]
Merge pull request #4082 from flx42/pin_base_docker_image
Pin the base image version for the GPU Dockerfile
Achal Dave [Wed, 4 May 2016 15:56:05 +0000 (11:56 -0400)]
Fix typo (indecies->indices)
Achal Dave [Wed, 4 May 2016 15:55:43 +0000 (11:55 -0400)]
Reformat to fit in 79 columns
Achal Dave [Wed, 4 May 2016 15:52:06 +0000 (11:52 -0400)]
Remove trailing spaces
Achal Dave [Wed, 4 May 2016 15:51:00 +0000 (11:51 -0400)]
Exit on error and report argument error details.
The statement 'exit' has no effect in Python scripts. Use 'sys.exit()'
instead.
ZhouYzzz [Wed, 4 May 2016 10:00:12 +0000 (18:00 +0800)]
Add test for attribute "phase" in python layer
Jon Long [Wed, 4 May 2016 06:32:22 +0000 (23:32 -0700)]
Merge pull request #3977 from tpwrules/master
Fix protobuf message generation
Jon Long [Wed, 4 May 2016 06:01:45 +0000 (23:01 -0700)]
Merge pull request #4040 from ebadawy/master
Read the data as a binary
Jeff Donahue [Wed, 4 May 2016 05:55:01 +0000 (22:55 -0700)]
Merge pull request #4056 from wk910930/fix-ReshapeParameter-example
Fix an error in the example of ReshapeParameter.
Jon Long [Wed, 4 May 2016 05:44:13 +0000 (22:44 -0700)]
Merge pull request #4065 from drewabbot/master
avoid non-integer array indices
Jon Long [Wed, 4 May 2016 05:26:18 +0000 (22:26 -0700)]
Merge pull request #4075 from szha/osx_makefile_fix
fix CUDA version detection in Makefile to accomodate OS X's grep
Felix Abecassis [Mon, 2 May 2016 23:20:00 +0000 (16:20 -0700)]
Pin the base image version for the GPU Dockerfile
The previous Dockerfile can break if image nvidia/cuda:cudnn is updated to any of the following:
- Ubuntu 16.04 LTS (already released)
- cuDNN v5 (soon)
- CUDA 8.0 (soon)
Andrew Tulloch [Sun, 1 May 2016 09:47:00 +0000 (02:47 -0700)]
Merge pull request #4071 from mnogu/optional-name
draw_net: accept prototxt without name
Sheng Zha [Sat, 30 Apr 2016 23:40:05 +0000 (16:40 -0700)]
fix grep in CUDA version detection to accomodate OSX's grep (and other grep that doesn't support \d extension)
Muneyuki Noguchi [Fri, 29 Apr 2016 02:04:02 +0000 (02:04 +0000)]
draw_net: accept prototxt without name
Fixes #3819
Evan Shelhamer [Thu, 28 Apr 2016 18:53:20 +0000 (11:53 -0700)]
Merge pull request #4069 from seanbell/pycaffe-boost-warnings-fix
suppress boost registration warnings in pycaffe (based on #3960)
Sean Bell [Thu, 28 Apr 2016 17:06:51 +0000 (13:06 -0400)]
Suppress boost registration warnings in pycaffe (Based on #3960)
Drew Abbot [Thu, 28 Apr 2016 06:25:09 +0000 (23:25 -0700)]
avoid non-integer array indices
Evan Shelhamer [Wed, 27 Apr 2016 21:29:18 +0000 (14:29 -0700)]
Merge pull request #4051 from samster25/master
fix typo in download script command cpp_classification
Kun Wang [Wed, 27 Apr 2016 13:09:31 +0000 (21:09 +0800)]
Fix an error in the example of ReshapeParameter.
* this small mistake may confuse newer.
Sammy Sidhu [Wed, 27 Apr 2016 10:05:30 +0000 (03:05 -0700)]
fixed typo in download script command cpp_classification
Jeff Donahue [Tue, 26 Apr 2016 23:11:12 +0000 (16:11 -0700)]
Merge pull request #4033 from HeGaoYuan/master
there is a very very small mistake in the LeNet MNIST tutorial
Evan Shelhamer [Mon, 25 Apr 2016 21:20:30 +0000 (14:20 -0700)]
Merge pull request #3731 from lukeyeager/lmdb-map-full
dynamically set LMDB map size (double when full)
ebadawy [Sun, 24 Apr 2016 18:24:41 +0000 (20:24 +0200)]
Read the data as a binary
Appending 'b' in the file mode as hashlib functions require to pass in bytes
HeGaoYuan [Sat, 23 Apr 2016 06:48:41 +0000 (14:48 +0800)]
Fixed #4029: test the network every 500 iterations, not 1000 iterations
Luke Yeager [Fri, 26 Feb 2016 04:14:02 +0000 (20:14 -0800)]
Update MNIST example to use new DB classes
Luke Yeager [Fri, 26 Feb 2016 04:02:25 +0000 (20:02 -0800)]
Print to stderr for example LMDB code
Luke Yeager [Fri, 26 Feb 2016 03:58:01 +0000 (19:58 -0800)]
Don't set map_size=1TB in util/db_lmdb
Instead, double the map size on the MDB_MAP_FULL exception.
Evan Shelhamer [Wed, 20 Apr 2016 21:57:30 +0000 (14:57 -0700)]
Merge pull request #4024 from achalddave/finetune-flickr-style-tutorial
[docs] explicitly point out -weights flag in fine-tuning tutorial
Evan Shelhamer [Wed, 20 Apr 2016 21:56:29 +0000 (14:56 -0700)]
Merge pull request #3749 from lukeyeager/bvlc/array_to_datum-default-label
Don't force datum.label=0 in array_to_datum
Achal Dave [Wed, 20 Apr 2016 21:34:29 +0000 (17:34 -0400)]
Explicitly point out -weights flag in tutorial
The -weights flag is somewhat easy to miss as it's only in one command,
but is the crucial thing that anyone searching for 'how to finetune' is
looking for. Hopefully this more clearly points out the '-weights' flag,
which might otherwise be overlooked in this tutorial.
Evan Shelhamer [Tue, 19 Apr 2016 00:51:52 +0000 (17:51 -0700)]
Merge pull request #4007 from lukeyeager/bvlc/docs-typo
[docs] typo in docs/installation.md
Luke Yeager [Mon, 18 Apr 2016 15:54:21 +0000 (08:54 -0700)]
Typo in docs/installation.md
ZhouYzzz [Fri, 15 Apr 2016 14:51:49 +0000 (22:51 +0800)]
Allow the python layer have attribute "phase"
Evan Shelhamer [Fri, 15 Apr 2016 08:02:02 +0000 (01:02 -0700)]
Merge pull request #3993 from shelhamer/fix-crop
Fix Crop layer dimension checking to only check cropped dimensions
Evan Shelhamer [Fri, 15 Apr 2016 05:31:38 +0000 (22:31 -0700)]
CropLayer: groom comments
Evan Shelhamer [Fri, 15 Apr 2016 05:16:07 +0000 (22:16 -0700)]
[fix] CropLayer: check dimension bounds only for cropped dimensions
check only the dimensions to be cropped for compatible sizes and offsets
Evan Shelhamer [Fri, 15 Apr 2016 05:56:37 +0000 (22:56 -0700)]
[test] CropLayer: test dimensions check to reveal bounds checking bug
Evan Shelhamer [Thu, 14 Apr 2016 09:19:27 +0000 (02:19 -0700)]
Merge pull request #3988 from shelhamer/install-docs
bring install docs up-to-date and better highlight included details
Evan Shelhamer [Thu, 14 Apr 2016 01:53:28 +0000 (18:53 -0700)]
[docs] install: include more lab tested hardware
Evan Shelhamer [Thu, 14 Apr 2016 01:52:50 +0000 (18:52 -0700)]
[docs] install: be more firm about compute capability >= 3.0
Evan Shelhamer [Wed, 13 Apr 2016 23:43:39 +0000 (16:43 -0700)]
[docs] install: include latest versions and platforms, highlight guides
Caffe runs on Ubuntu, OS X, and RHEL (+ company) in master
with branches for OpenCL and Windows.
Docker is a nice route to out-of-the-box brewing.
Evan Shelhamer [Wed, 13 Apr 2016 23:40:30 +0000 (16:40 -0700)]
[docs] install: CUDA 7+ and cuDNN v4 compatible
Latest CUDA versions are all compatible, and
Caffe has been compatible with cuDNN v4 since PR #3439
Evan Shelhamer [Wed, 13 Apr 2016 01:18:26 +0000 (18:18 -0700)]
Merge pull request #3982 from mnogu/fix-typo-model-option
[docs] typo in help text for "-model" option
Muneyuki Noguchi [Tue, 12 Apr 2016 14:19:27 +0000 (23:19 +0900)]
Fix typo in help text for "-model" option
Thomas [Mon, 11 Apr 2016 17:52:34 +0000 (12:52 -0500)]
Fix protobuf message generation
The latest versions of protobuf do not reveal empty message fields with dir(). This uses the documented way of determining all of a message's fields and so is compatible with past and future versions of protobuf.
Jeff Donahue [Fri, 8 Apr 2016 22:07:20 +0000 (15:07 -0700)]
Merge pull request #3937 from emaggiori/exp
Issue with the exp layer when the base is e
emmanuel maggiori [Fri, 8 Apr 2016 08:25:12 +0000 (10:25 +0200)]
Solving issue with exp layer with base e
Jeff Donahue [Tue, 5 Apr 2016 17:15:16 +0000 (10:15 -0700)]
Merge pull request #3942 from jeffdonahue/propagate-down-true
Set propagate_down=true to force backprop to a particular bottom
Jeff Donahue [Wed, 27 Jan 2016 20:55:41 +0000 (12:55 -0800)]
Net: setting `propagate_down: true` forces backprop
Jeff Donahue [Mon, 4 Apr 2016 18:36:15 +0000 (11:36 -0700)]
test_net.cpp: add TestForcePropagateDown