platform/upstream/caffeonacl.git
9 years agotest equivalence of solving with accumulating gradients
Evan Shelhamer [Fri, 22 May 2015 01:14:16 +0000 (18:14 -0700)]
test equivalence of solving with accumulating gradients

Compare the parameters after solving with a given batch size and the
halved batch size + two iter accumulation of gradients equivalent.

Note: the test net dummy data layer now makes constant data and random
gaussian targets. This assures the standard and gradient accumulation
cases check the same data. Otherwise the difference in batch sizes
causes different orders of random number draws.

9 years agoadjust local learning rate and decay according to gradient accumulation
Evan Shelhamer [Fri, 22 May 2015 00:06:42 +0000 (17:06 -0700)]
adjust local learning rate and decay according to gradient accumulation

Divide local rate by `iter_size` to normalize the gradient according to
the full minibatch size and not only the computational batch size.

Multiply the local decay by `iter_size` to counter the division of the
local learning rate since the decay is multiplied by the rate in the
update equation.

9 years agoaccumulate gradients in cudnn conv layer
Jonathan L Long [Sun, 14 Sep 2014 00:41:59 +0000 (17:41 -0700)]
accumulate gradients in cudnn conv layer

9 years agoaccumulate gradients in (de)conv layers
Jonathan L Long [Wed, 31 Dec 2014 06:29:35 +0000 (22:29 -0800)]
accumulate gradients in (de)conv layers

9 years agoaccumulate gradients in inner product layer
Sergio [Sat, 27 Sep 2014 06:03:26 +0000 (23:03 -0700)]
accumulate gradients in inner product layer

9 years agozero-init param diffs in gradient checker
Jonathan L Long [Wed, 31 Dec 2014 06:52:07 +0000 (22:52 -0800)]
zero-init param diffs in gradient checker

9 years agozero-init param diffs and accumulate gradients
Jonathan L Long [Tue, 12 Aug 2014 04:38:59 +0000 (21:38 -0700)]
zero-init param diffs and accumulate gradients

(With layers whose backward accumulates gradients), this effectively
decouples the computational batch from the SGD minibatch. Each
iteration accumulates gradients over iter_size batches, then parameters
are updated.

9 years agoMerge pull request #2518 from shelhamer/dedup_solvers
Jeff Donahue [Wed, 27 May 2015 20:42:47 +0000 (13:42 -0700)]
Merge pull request #2518 from shelhamer/dedup_solvers

Deduplicate solver regularization, logging, and local rates and decays

9 years agoSolver::MakeUpdate() -> Solver::ApplyUpdate
Evan Shelhamer [Wed, 27 May 2015 19:24:06 +0000 (12:24 -0700)]
Solver::MakeUpdate() -> Solver::ApplyUpdate

Designate `Solver::ApplyUpdate()` as the core method to compute
and apply parameter updates given the current state of the Net.

Make `Solver::ComputeUpdateValue()` a subordinate call overloaded by the
`SGDSolver`s to take care of optimization algorithm details.

9 years agodeduplicate decay and local rate in solver updates
Evan Shelhamer [Thu, 21 May 2015 23:34:43 +0000 (16:34 -0700)]
deduplicate decay and local rate in solver updates

9 years agoRefactor solvers regularization and logging code
Cyprien Noel [Tue, 19 May 2015 01:30:00 +0000 (18:30 -0700)]
Refactor solvers regularization and logging code

9 years agoMerge pull request #1946 from nickcarlevaris/msra_init
Evan Shelhamer [Tue, 26 May 2015 20:13:01 +0000 (13:13 -0700)]
Merge pull request #1946 from nickcarlevaris/msra_init

  Add MSRAFiller, an Xavier-like filler designed for use with ReLUs

9 years agoinclude comment on Saxe and sqrt(2) scaling factor
Evan Shelhamer [Tue, 26 May 2015 19:39:14 +0000 (12:39 -0700)]
include comment on Saxe and sqrt(2) scaling factor

although different and independent, the derivation of Saxe et
al. with regards to the scaling factor might be of interest.

9 years agoAdded MSRAFiller, an Xavier-like filler designed for use with ReLUs
Nick Carlevaris-Bianco [Mon, 16 Feb 2015 05:19:43 +0000 (15:49 +1030)]
Added MSRAFiller, an Xavier-like filler designed for use with ReLUs

...instead of tanh. Based on paper: He et al, "Delving Deep into
Rectifiers: Surpassing Human-Level Performance on ImageNet
Classification," 2015.

- add VarianceNorm option to FillerParameters which allows one to
  normalize by fan_in, fan_out or their average.
- update XavierFiller to use the VarianceNorm option (default behavior
  unchanged).
- add tests for MSRAFiller and XavierFiller.

9 years agoMerge pull request #2486 from tiangolo/ipython-notebook-v4
Evan Shelhamer [Fri, 22 May 2015 20:27:07 +0000 (13:27 -0700)]
Merge pull request #2486 from tiangolo/ipython-notebook-v4

Update IPython Notebooks to version 4, fix #2485

9 years agoUpdate python/requirements.txt to have ipython>=3.0.0
Sebastián Ramírez [Fri, 22 May 2015 01:19:59 +0000 (20:19 -0500)]
Update python/requirements.txt to have ipython>=3.0.0

9 years agoMerge pull request #2488 from kibum14/master
Evan Shelhamer [Tue, 19 May 2015 18:11:38 +0000 (11:11 -0700)]
Merge pull request #2488 from kibum14/master

[docs] fix typos

9 years agofix typos in docs
Kibum Bae [Tue, 19 May 2015 13:43:49 +0000 (22:43 +0900)]
fix typos in docs

fix typos in install_osx.md and performance_hardware.md

9 years agoMerge pull request #2484 from ronghanghu/fix-caffe-test
Evan Shelhamer [Tue, 19 May 2015 05:46:56 +0000 (22:46 -0700)]
Merge pull request #2484 from ronghanghu/fix-caffe-test

[bug] fix blob_loss_weights index in test() in caffe.cpp

9 years agoMerge pull request #2482 from longjon/clean-message-comments
Evan Shelhamer [Mon, 18 May 2015 18:55:42 +0000 (11:55 -0700)]
Merge pull request #2482 from longjon/clean-message-comments

Clean up redundant protobuf message comments

9 years agoUpdate IPython Notebooks to version 4
Sebastián Ramírez [Mon, 18 May 2015 15:21:44 +0000 (10:21 -0500)]
Update IPython Notebooks to version 4

9 years agofix blob_loss_weights index in test() in caffe.cpp
Ronghang Hu [Mon, 18 May 2015 12:27:19 +0000 (20:27 +0800)]
fix blob_loss_weights index in test() in caffe.cpp

Correct the index for blob_loss_weights during output. Previously it was set to test_score index by mistake.

9 years agoclean up redundant message comments
Jonathan L Long [Mon, 18 May 2015 07:45:26 +0000 (00:45 -0700)]
clean up redundant message comments

9 years agoMerge pull request #2466 from ducha-aiki/mvn-less
Jeff Donahue [Sat, 16 May 2015 19:02:59 +0000 (12:02 -0700)]
Merge pull request #2466 from ducha-aiki/mvn-less

Remove unnecessary variance computation from backward in MVN layer

9 years agoMerge pull request #2095 from mtamburrano/skip_propagate_down_param
Jeff Donahue [Fri, 15 May 2015 19:51:16 +0000 (12:51 -0700)]
Merge pull request #2095 from mtamburrano/skip_propagate_down_param

Added param skip_propagate_down to LayerParameter

9 years agoMerge pull request #2467 from MartinThoma/moose
Jon Long [Fri, 15 May 2015 17:43:30 +0000 (10:43 -0700)]
Merge pull request #2467 from MartinThoma/moose

Python: Formatted docstrings to numpydoc (Take, Give -> Parameters, Returns)

9 years agoPython: Formatted docstrings to numpydoc (Take, Give -> Parameters, Returns)
Martin Thoma [Fri, 15 May 2015 14:06:49 +0000 (16:06 +0200)]
Python: Formatted docstrings to numpydoc (Take, Give -> Parameters, Returns)

9 years agoRemove unnecessary variance computation from backward in MVN layer
Dmytro Mishkin [Fri, 15 May 2015 13:14:02 +0000 (16:14 +0300)]
Remove unnecessary variance computation from backward in MVN layer

9 years agoAdded "propagate_down" param to LayerParameter
manuele [Fri, 15 May 2015 09:17:00 +0000 (11:17 +0200)]
Added "propagate_down" param to LayerParameter

9 years agoMerge pull request #2274 from Ashwani001/patch-1
Jeff Donahue [Fri, 15 May 2015 07:10:30 +0000 (00:10 -0700)]
Merge pull request #2274 from Ashwani001/patch-1

Update generate_sample_data.py

9 years agoUpdate generate_sample_data.py
Ashwani001 [Fri, 15 May 2015 02:33:36 +0000 (10:33 +0800)]
Update generate_sample_data.py

made code more clearer and more concise

9 years agoMerge pull request #2201 from jeffdonahue/tutorial-fixes
Evan Shelhamer [Fri, 15 May 2015 01:42:02 +0000 (18:42 -0700)]
Merge pull request #2201 from jeffdonahue/tutorial-fixes

Update docs for ND blobs (#1970) and layer type is a string (#1694)

9 years agoUpdate docs for ND blobs (#1970) and layer type is a string (#1694)
Jeff Donahue [Thu, 26 Mar 2015 01:45:29 +0000 (18:45 -0700)]
Update docs for ND blobs (#1970) and layer type is a string (#1694)

9 years agoMerge pull request #2217 from jeffdonahue/ssafar-reshape-rebase
Jeff Donahue [Fri, 15 May 2015 01:36:22 +0000 (18:36 -0700)]
Merge pull request #2217 from jeffdonahue/ssafar-reshape-rebase

Rebase @ssafar's ReshapeLayer

9 years agoAdd ReshapeParameter axis and num_axes to reshape only a particular span
Jeff Donahue [Thu, 26 Mar 2015 08:13:18 +0000 (01:13 -0700)]
Add ReshapeParameter axis and num_axes to reshape only a particular span
of the input shape

9 years agobasic tests (Forward, Gradient) for ReshapeLayer
Jeff Donahue [Thu, 26 Mar 2015 09:25:48 +0000 (02:25 -0700)]
basic tests (Forward, Gradient) for ReshapeLayer

9 years agoReshapeLayer fixups for ND blobs
Jeff Donahue [Thu, 26 Mar 2015 00:44:37 +0000 (17:44 -0700)]
ReshapeLayer fixups for ND blobs

9 years agoAdded a Reshape layer for copying-free modification of blob dimensions.
Simon Safar [Thu, 16 Oct 2014 03:15:14 +0000 (20:15 -0700)]
Added a Reshape layer for copying-free modification of blob dimensions.

9 years agoMerge pull request #2177 from pgao/spp_layer
Jeff Donahue [Fri, 15 May 2015 01:17:11 +0000 (18:17 -0700)]
Merge pull request #2177 from pgao/spp_layer

Spatial Pyramid Pooling Layer

9 years agoSpatial Pyramid Pooling Layer
PETER_GAO [Sat, 21 Mar 2015 23:00:05 +0000 (16:00 -0700)]
Spatial Pyramid Pooling Layer

9 years agoMerge pull request #2115 from longjon/bogus-cross-entropy-gpu
Jeff Donahue [Thu, 14 May 2015 23:11:06 +0000 (16:11 -0700)]
Merge pull request #2115 from longjon/bogus-cross-entropy-gpu

Remove bogus implementation of SigmoidCrossEntropyLossLayer's Forward_gpu

9 years agoremove bogus implementation of SigmoidCrossEntropyLossLayer::Forward_gpu
Jonathan L Long [Fri, 13 Mar 2015 00:59:28 +0000 (17:59 -0700)]
remove bogus implementation of SigmoidCrossEntropyLossLayer::Forward_gpu

It was a verbatim copy of Forward_cpu; there is no proper GPU
implementation.

9 years agoMerge pull request #1969 from tnarihi/fix-empty-param_name
Jeff Donahue [Thu, 14 May 2015 22:41:57 +0000 (15:41 -0700)]
Merge pull request #1969 from tnarihi/fix-empty-param_name

Fix incorrectly storing empty param_name to param_names_index_

9 years agoMerge pull request #2168 from longjon/spurious-net-includes
Jeff Donahue [Thu, 14 May 2015 22:14:05 +0000 (15:14 -0700)]
Merge pull request #2168 from longjon/spurious-net-includes

Remove spurious inclusions of net.hpp

9 years agoMerge pull request #2165 from longjon/auto-reshape
Jeff Donahue [Thu, 14 May 2015 22:10:59 +0000 (15:10 -0700)]
Merge pull request #2165 from longjon/auto-reshape

Always call Layer::Reshape in Layer::Forward

9 years agoMerge pull request #2072 from jeffdonahue/final-snapshot-off-by-one
Jon Long [Thu, 14 May 2015 22:06:56 +0000 (15:06 -0700)]
Merge pull request #2072 from jeffdonahue/final-snapshot-off-by-one

Bugfix: final snapshot iter number is off by one

9 years agoMerge pull request #2456 from longjon/python-layer-object
Jeff Donahue [Thu, 14 May 2015 22:06:04 +0000 (15:06 -0700)]
Merge pull request #2456 from longjon/python-layer-object

Use bp::object instead of PyObject* for self in Python layer

9 years agoMerge pull request #2457 from longjon/superfluous-destructors
Jeff Donahue [Thu, 14 May 2015 21:57:52 +0000 (14:57 -0700)]
Merge pull request #2457 from longjon/superfluous-destructors

Remove superfluous empty destructors

9 years agoremove superfluous empty destructors
Jonathan L Long [Thu, 14 May 2015 05:08:57 +0000 (22:08 -0700)]
remove superfluous empty destructors

The removed definitions do nothing; these classes already have virtual
destructors inherited from their respective base classes.

9 years ago[pycaffe] use bp::object instead of PyObject* for self in Python layer
Takuya Narihira [Thu, 14 May 2015 04:16:28 +0000 (21:16 -0700)]
[pycaffe] use bp::object instead of PyObject* for self in Python layer

This simply allows direct use of the nicer bp::object interface.

9 years agoMerge pull request #2321 from nickcarlevaris/contrastive_loss_fix
Evan Shelhamer [Tue, 12 May 2015 20:52:48 +0000 (13:52 -0700)]
Merge pull request #2321 from nickcarlevaris/contrastive_loss_fix

Fixed contrastive loss layer to be the same as proposed in Hadsell et al 2006

9 years agoMerge pull request #2441 from gustavla/py3-fix
Jon Long [Mon, 11 May 2015 23:34:33 +0000 (16:34 -0700)]
Merge pull request #2441 from gustavla/py3-fix

Fixed wrong io import in Python 3

9 years agoMerge pull request #2443 from MartinThoma/master
Evan Shelhamer [Mon, 11 May 2015 22:20:06 +0000 (15:20 -0700)]
Merge pull request #2443 from MartinThoma/master

python: PEP8; changed docstring documentation style to NumPyDoc style

9 years agopython: PEP8; changed docstring documentation style to NumPyDoc style
Martin Thoma [Mon, 11 May 2015 20:35:48 +0000 (22:35 +0200)]
python: PEP8; changed docstring documentation style to NumPyDoc style

9 years agoThis imports the wrong io module in Python 3.
Gustav Larsson [Mon, 11 May 2015 16:54:45 +0000 (11:54 -0500)]
This imports the wrong io module in Python 3.

The Python standard lib has a module called io, so instead of Python 3
throwing an error, it imports the wrong module without complaining.

9 years agoMerge pull request #2426 from longjon/check-blob-overflow
Jeff Donahue [Thu, 7 May 2015 00:56:57 +0000 (17:56 -0700)]
Merge pull request #2426 from longjon/check-blob-overflow

Check that count does not overflow in Blob::Reshape

9 years agocheck that count_ does not overflow in Blob::Reshape
Jonathan L Long [Thu, 7 May 2015 00:40:12 +0000 (17:40 -0700)]
check that count_ does not overflow in Blob::Reshape

9 years agoMerge pull request #2414 from tnarihi/fix-prelu-redanduncy
Jeff Donahue [Tue, 5 May 2015 18:39:45 +0000 (11:39 -0700)]
Merge pull request #2414 from tnarihi/fix-prelu-redanduncy

Fix #2406: wrong thread blocks setting for PReLU

9 years agoModify for better readability regarding temporary bufffer for backward
Takuya Narihira [Mon, 4 May 2015 18:45:33 +0000 (11:45 -0700)]
Modify for better readability regarding temporary bufffer for backward
computation

9 years agoFix redundancy of parameter backward computation
Takuya Narihira [Mon, 4 May 2015 18:44:44 +0000 (11:44 -0700)]
Fix redundancy of parameter backward computation

9 years agoAdded support for original implementation, using (margin - d^2), through the
Nick Carlevaris-Bianco [Mon, 4 May 2015 02:11:44 +0000 (11:41 +0930)]
Added support for original implementation, using (margin - d^2), through the
legacy_version parameter.

9 years agoMerge pull request #2369 from jeffdonahue/makefile-link-bugfix
Jeff Donahue [Thu, 30 Apr 2015 22:47:03 +0000 (15:47 -0700)]
Merge pull request #2369 from jeffdonahue/makefile-link-bugfix

Makefile bugfix: debug/release dir link not toggled properly

9 years agoMakefile bugfix: OTHER_BUILD_DIR name set incorrectly when empty due to
Jeff Donahue [Sun, 26 Apr 2015 03:57:34 +0000 (20:57 -0700)]
Makefile bugfix: OTHER_BUILD_DIR name set incorrectly when empty due to
lazy variable expansion when using the `?=` operator -- change them
to explicit empty string checks with simple assignment operator `:=`.

9 years agoMerge pull request #2393 from Nerei/feature/minor_cmake_fix
Jon Long [Thu, 30 Apr 2015 03:10:14 +0000 (20:10 -0700)]
Merge pull request #2393 from Nerei/feature/minor_cmake_fix

Little CMake fix (can be reviewed and merged quickly)

9 years agominor cmake fix - now Caffe complains when cmake is executed if glog/gflags are not...
Anatoly Baksheev [Wed, 29 Apr 2015 18:53:51 +0000 (21:53 +0300)]
minor cmake fix - now Caffe complains when cmake is executed if glog/gflags are not found.

9 years agoMerge pull request #2365 from wakamori/fix_typo_in_gnuplot_example
Jeff Donahue [Tue, 28 Apr 2015 07:59:32 +0000 (00:59 -0700)]
Merge pull request #2365 from wakamori/fix_typo_in_gnuplot_example

Typo fix: the label titles in plot_log.gnuplot.example

9 years agoImport Pandas in HDF5 IPython notebook. Fix for issue BVLC/caffe#2247
Sebastián Ramírez [Fri, 3 Apr 2015 01:01:24 +0000 (20:01 -0500)]
Import Pandas in HDF5 IPython notebook. Fix for issue BVLC/caffe#2247

9 years agoMerge pull request #2370 from gdh1995/master
Jon Long [Sun, 26 Apr 2015 22:02:26 +0000 (15:02 -0700)]
Merge pull request #2370 from gdh1995/master

move CPU_ONLY to right place

9 years agoNet::Update: CPU_ONLY is in wrong place
gdh1995 [Sun, 26 Apr 2015 14:52:41 +0000 (22:52 +0800)]
Net::Update: CPU_ONLY is in wrong place

9 years agofix a typo that GFLAGS_GFLAGS_H_ -> GFLAGS_GFAGS_H_
gdh1995 [Sun, 26 Apr 2015 14:51:21 +0000 (22:51 +0800)]
fix a typo that GFLAGS_GFLAGS_H_ -> GFLAGS_GFAGS_H_

9 years agofix typo: swap the titles of xlabel and ylabel
Takuma Wakamori [Sat, 25 Apr 2015 08:09:43 +0000 (17:09 +0900)]
fix typo: swap the titles of xlabel and ylabel

9 years agoMerge pull request #2352 from pgao/rcnn-fix
Jon Long [Sat, 25 Apr 2015 03:26:06 +0000 (20:26 -0700)]
Merge pull request #2352 from pgao/rcnn-fix

Fix RCNN model fetching script

9 years agoMerge pull request #2330 from flx42/include_config_file_check
Jonathan L Long [Sat, 25 Apr 2015 02:55:31 +0000 (19:55 -0700)]
Merge pull request #2330 from flx42/include_config_file_check

9 years agoclarify Makefile.config check
Jonathan L Long [Sat, 25 Apr 2015 02:52:33 +0000 (19:52 -0700)]
clarify Makefile.config check

9 years agoFix RCNN model fetching script
PETER_GAO [Wed, 22 Apr 2015 18:33:41 +0000 (11:33 -0700)]
Fix RCNN model fetching script

9 years agoadded epsilon to prevent possible division by zero in gradient calculation
Nick Carlevaris-Bianco [Tue, 21 Apr 2015 07:44:47 +0000 (17:14 +0930)]
added epsilon to prevent possible division by zero in gradient calculation

9 years agoAbort Makefile parsing if the configuration file cannot be found.
Felix Abecassis [Fri, 17 Apr 2015 22:53:12 +0000 (15:53 -0700)]
Abort Makefile parsing if the configuration file cannot be found.

9 years agoset default DISTRIBUTE_DIR -- fix #2328
Evan Shelhamer [Fri, 17 Apr 2015 19:26:48 +0000 (12:26 -0700)]
set default DISTRIBUTE_DIR -- fix #2328

9 years agoFixed contrastive loss layer to be the same as proposed in Hadsell et al 2006
Nick Carlevaris-Bianco [Fri, 17 Apr 2015 04:46:26 +0000 (14:16 +0930)]
Fixed contrastive loss layer to be the same as proposed in Hadsell et al 2006

9 years agoMerge pull request #2295 from akiomik/patch-1
Evan Shelhamer [Fri, 10 Apr 2015 18:01:33 +0000 (11:01 -0700)]
Merge pull request #2295 from akiomik/patch-1

[docs] fix brew command for OS X install

9 years agoimproved installation for osx
Akiomi KAMAKURA [Fri, 10 Apr 2015 06:48:41 +0000 (15:48 +0900)]
improved installation for osx

9 years agoMerge pull request #2287 from eerwitt/web-demo-import
Evan Shelhamer [Wed, 8 Apr 2015 22:44:36 +0000 (15:44 -0700)]
Merge pull request #2287 from eerwitt/web-demo-import

[example] change `import Image` for forward compatibility

9 years agoChanging Image import to be imported from PIL.
e3 [Wed, 8 Apr 2015 22:00:36 +0000 (15:00 -0700)]
Changing Image import to be imported from PIL.

9 years agoUpdate generate_sample_data.py
Ashwani001 [Tue, 7 Apr 2015 09:44:01 +0000 (17:44 +0800)]
Update generate_sample_data.py

Changed os.path.dirname(__file__) as it was returning empty...

9 years ago[docs] switch lmdb url for manual install, tweak formatting
Evan Shelhamer [Sun, 5 Apr 2015 18:47:49 +0000 (11:47 -0700)]
[docs] switch lmdb url for manual install, tweak formatting

9 years agoMerge pull request #2224 from small-yellow-duck/master
Evan Shelhamer [Tue, 31 Mar 2015 18:14:30 +0000 (11:14 -0700)]
Merge pull request #2224 from small-yellow-duck/master

[build] check if CPU_ONLY is set when determining CUDA version

9 years agoMerge pull request #2231 from tnarihi/fix-travis-miniconda
Evan Shelhamer [Tue, 31 Mar 2015 18:13:20 +0000 (11:13 -0700)]
Merge pull request #2231 from tnarihi/fix-travis-miniconda

Fix Travis: no need to remove libm in new Miniconda

9 years agoFix: libm.* doesn not exist
Takuya Narihira [Tue, 31 Mar 2015 02:52:43 +0000 (19:52 -0700)]
Fix: libm.* doesn not exist

9 years agoCheck if CPU_ONLY is set when determining CUDA version
small-yellow-duck [Sun, 29 Mar 2015 21:30:21 +0000 (14:30 -0700)]
Check if CPU_ONLY is set when determining CUDA version

Previously, CUDA_VERSION would appear to be < 7 if there was no CUDA
installed, and that would generate the wrong C++ flags for compiling on
recent OSX versions. Instead, skip the CUDA version check if CPU_ONLY is
set. This change only affects CPU_ONLY installations.

9 years agoMerge pull request #2192 from lukeyeager/remove-scikit-learn
Evan Shelhamer [Fri, 27 Mar 2015 22:58:14 +0000 (15:58 -0700)]
Merge pull request #2192 from lukeyeager/remove-scikit-learn

Remove scikit-learn dependency -- the need is noted in the relevant example

9 years agoMerge pull request #2199 from lukeyeager/downgrade-pillow
Jon Long [Fri, 27 Mar 2015 22:27:43 +0000 (15:27 -0700)]
Merge pull request #2199 from lukeyeager/downgrade-pillow

Downgrade Pillow pip requirement

9 years agoMerge pull request #2211 from nsubtil/fix-cudnn-algo
Evan Shelhamer [Thu, 26 Mar 2015 23:50:52 +0000 (16:50 -0700)]
Merge pull request #2211 from nsubtil/fix-cudnn-algo

Fallback to different cuDNN algorithm when under memory pressure; fix #2197

9 years agoFallback to different cuDNN algorithm when under memory pressure
Nuno Subtil [Tue, 24 Mar 2015 22:48:16 +0000 (15:48 -0700)]
Fallback to different cuDNN algorithm when under memory pressure

CUDNN_CONVOLUTION_FWD_PREFER_FASTEST requires a lot of GPU memory, which may
not always be available. Add a fallback path that uses
CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM when the allocation fails.

9 years agoDowngrade Pillow pip requirement
Luke Yeager [Wed, 25 Mar 2015 19:42:17 +0000 (12:42 -0700)]
Downgrade Pillow pip requirement

2.7.0 isn't really necessary - 2.3.0 is sufficient. This is the version
available on Ubuntu 14.04 via apt-get, and seems to be a reasonable lowest
common denominator in general.

http://pillow.readthedocs.org/installation.html#old-versions

9 years agoMerge pull request #2160 from TorosFanny/master
Jeff Donahue [Wed, 25 Mar 2015 17:42:22 +0000 (10:42 -0700)]
Merge pull request #2160 from TorosFanny/master

change resorce to resource

9 years agoAdd note in example about installing scikit-learn
Luke Yeager [Wed, 25 Mar 2015 01:15:08 +0000 (18:15 -0700)]
Add note in example about installing scikit-learn

9 years agoRemove scikit-learn dependency
Luke Yeager [Wed, 25 Mar 2015 00:42:56 +0000 (17:42 -0700)]
Remove scikit-learn dependency

9 years agoMerge pull request #2038 from shelhamer/cudnn-r2
Evan Shelhamer [Tue, 24 Mar 2015 21:49:31 +0000 (14:49 -0700)]
Merge pull request #2038 from shelhamer/cudnn-r2

cuDNN v2

9 years agonote cuDNN v2 convolutional TODOs
Evan Shelhamer [Tue, 24 Mar 2015 21:17:57 +0000 (14:17 -0700)]
note cuDNN v2 convolutional TODOs

9 years agocuDNN pooling can pad now
Evan Shelhamer [Tue, 17 Feb 2015 01:18:13 +0000 (17:18 -0800)]
cuDNN pooling can pad now