platform/upstream/caffeonacl.git
10 years agoCorrect reference to lenet_train_test.prototxt
Daniel Golden [Wed, 3 Sep 2014 20:18:48 +0000 (13:18 -0700)]
Correct reference to lenet_train_test.prototxt

Not lenet.prototxt

10 years agoPoint to local file, not github file
Daniel Golden [Wed, 3 Sep 2014 20:01:29 +0000 (13:01 -0700)]
Point to local file, not github file

This makes the readme file more consistent

10 years agoUpdate paths
Daniel Golden [Wed, 3 Sep 2014 20:00:58 +0000 (13:00 -0700)]
Update paths

10 years agoMerge pull request #955 from kloudkl/data-layers
Evan Shelhamer [Wed, 3 Sep 2014 21:04:06 +0000 (14:04 -0700)]
Merge pull request #955 from kloudkl/data-layers

Further simplify the data layers and extend the MemoryDataLayer.

- design BaseDataLayer and BasePrefectingDataLayer for simplification
- refactor DataLayer, ImageDataLayer, and WindowDataLayer
- pull up `transform_param` into layer message for de-duplication
- add transformation to MemoryDataLayer

10 years ago[example] upgrade fine-tuning example to new transformation param
Evan Shelhamer [Wed, 3 Sep 2014 20:58:00 +0000 (13:58 -0700)]
[example] upgrade fine-tuning example to new transformation param

10 years agoMerge pull request #973 from shelhamer/tutorial-docs
Evan Shelhamer [Wed, 3 Sep 2014 20:27:48 +0000 (13:27 -0700)]
Merge pull request #973 from shelhamer/tutorial-docs

Document Caffe by a tour of key subjects + doxygen dev documentation

10 years ago[docs] configure doxygen + docs script for docs/doxygen site output
Evan Shelhamer [Wed, 3 Sep 2014 20:22:09 +0000 (13:22 -0700)]
[docs] configure doxygen + docs script for docs/doxygen site output

10 years agoupdate doxygen config to stop warnings
Evan Shelhamer [Wed, 3 Sep 2014 19:59:55 +0000 (12:59 -0700)]
update doxygen config to stop warnings

10 years ago[docs] suggest the CVPR14 deep learning tutorial for nice contrast
Evan Shelhamer [Wed, 3 Sep 2014 19:46:47 +0000 (12:46 -0700)]
[docs] suggest the CVPR14 deep learning tutorial for nice contrast

10 years ago[docs] draft data
Evan Shelhamer [Wed, 3 Sep 2014 19:25:52 +0000 (12:25 -0700)]
[docs] draft data

10 years agowrap up solver.md -- add update info for all solvers with citations;
Jeff Donahue [Wed, 3 Sep 2014 07:44:23 +0000 (00:44 -0700)]
wrap up solver.md -- add update info for all solvers with citations;
rules of thumb for SGD

10 years agonet.hpp: Doxygen-format docs
Jeff Donahue [Sat, 30 Aug 2014 22:30:48 +0000 (15:30 -0700)]
net.hpp: Doxygen-format docs

10 years agosolver.hpp: add \briefs
Jeff Donahue [Sat, 30 Aug 2014 22:12:39 +0000 (15:12 -0700)]
solver.hpp: add \briefs

10 years agosyncedmem.hpp: \brief and todo
Jeff Donahue [Sat, 30 Aug 2014 22:06:16 +0000 (15:06 -0700)]
syncedmem.hpp: \brief and todo

10 years agoblob.hpp: a little Doxygen-style documentation
Jeff Donahue [Sat, 30 Aug 2014 22:02:22 +0000 (15:02 -0700)]
blob.hpp: a little Doxygen-style documentation

10 years agofiller.hpp: add brief filler descriptions
Jeff Donahue [Sat, 30 Aug 2014 21:46:13 +0000 (14:46 -0700)]
filler.hpp: add brief filler descriptions

10 years agovision_layers.hpp: Doxygen \brief & TODO stubs.
Jeff Donahue [Sat, 30 Aug 2014 05:57:27 +0000 (22:57 -0700)]
vision_layers.hpp: Doxygen \brief & TODO stubs.

10 years agodata_layers: Doxygen \brief & TODO stubs.
Jeff Donahue [Sat, 30 Aug 2014 05:48:11 +0000 (22:48 -0700)]
data_layers: Doxygen \brief & TODO stubs.

10 years agocommon_layers.hpp: Doxygen \brief & TODO stubs.
Jeff Donahue [Sat, 30 Aug 2014 04:53:12 +0000 (21:53 -0700)]
common_layers.hpp: Doxygen \brief & TODO stubs.

10 years agoneuron_layers.hpp: Doxygen-style documentation
Jeff Donahue [Thu, 28 Aug 2014 09:36:35 +0000 (02:36 -0700)]
neuron_layers.hpp: Doxygen-style documentation

10 years agoloss_layers.hpp: Doxygen-style documentation
Jeff Donahue [Thu, 28 Aug 2014 11:17:43 +0000 (04:17 -0700)]
loss_layers.hpp: Doxygen-style documentation

10 years agolayer.hpp: Doxygen-style documentation
Jeff Donahue [Thu, 28 Aug 2014 07:24:12 +0000 (00:24 -0700)]
layer.hpp: Doxygen-style documentation

10 years ago.Doxyfile: don't warn if undocumented (maybe someday...)
Jeff Donahue [Fri, 29 Aug 2014 10:31:08 +0000 (03:31 -0700)]
.Doxyfile: don't warn if undocumented (maybe someday...)

10 years ago.Doxyfile: modify to generate C++ docs, excluding tests
Jeff Donahue [Thu, 28 Aug 2014 05:39:51 +0000 (22:39 -0700)]
.Doxyfile: modify to generate C++ docs, excluding tests

10 years ago.gitignore doxygen-generated documentation
Jeff Donahue [Thu, 28 Aug 2014 08:48:49 +0000 (01:48 -0700)]
.gitignore doxygen-generated documentation

10 years agoadd "make {docs,doxygen}" targets to build doxygen-generated docs
Jeff Donahue [Thu, 28 Aug 2014 08:44:02 +0000 (01:44 -0700)]
add "make {docs,doxygen}" targets to build doxygen-generated docs

10 years agoadd .Doxyfile: the default Doxygen config file from `doxygen -g`
Jeff Donahue [Thu, 28 Aug 2014 05:21:49 +0000 (22:21 -0700)]
add .Doxyfile: the default Doxygen config file from `doxygen -g`

10 years ago[wip] vision layers, start convolution
Jeff Donahue [Thu, 28 Aug 2014 02:38:13 +0000 (19:38 -0700)]
[wip] vision layers, start convolution

10 years agouse kramdown for markdown syntax; add mathjax
Jeff Donahue [Thu, 28 Aug 2014 04:16:18 +0000 (21:16 -0700)]
use kramdown for markdown syntax; add mathjax

10 years ago[docs] add note on Caffe convolution
Evan Shelhamer [Wed, 3 Sep 2014 17:58:25 +0000 (10:58 -0700)]
[docs] add note on Caffe convolution

10 years ago[docs] draft tutorial subjects
Evan Shelhamer [Sun, 24 Aug 2014 02:22:09 +0000 (19:22 -0700)]
[docs] draft tutorial subjects

10 years ago[docs] skeleton documentation subjects
Evan Shelhamer [Sun, 24 Aug 2014 01:19:10 +0000 (18:19 -0700)]
[docs] skeleton documentation subjects

10 years agoInitialize the transformer rng in the base data layer
Kai Li [Wed, 3 Sep 2014 07:13:26 +0000 (15:13 +0800)]
Initialize the transformer rng in the base data layer

10 years agoCorrect the datum size checking conditions of the data layers
Kai Li [Wed, 3 Sep 2014 06:48:02 +0000 (14:48 +0800)]
Correct the datum size checking conditions of the data layers

10 years agoAdd and transform Datum vector in the MemeoryDataLayer
Kai Li [Tue, 2 Sep 2014 06:43:58 +0000 (14:43 +0800)]
Add and transform Datum vector in the MemeoryDataLayer

10 years agoPlace InternalThreadEntry lower in the {,Image,Window}DataLayer.cpp
Kai Li [Tue, 2 Sep 2014 05:53:18 +0000 (13:53 +0800)]
Place InternalThreadEntry lower in the {,Image,Window}DataLayer.cpp

10 years agoAdd leveldb header back to util/io.cpp
Kai Li [Tue, 2 Sep 2014 05:44:59 +0000 (13:44 +0800)]
Add leveldb header back to util/io.cpp

10 years agoRemove OpenCV stuffs from the memory data layer and io utils
Kai Li [Tue, 2 Sep 2014 03:50:08 +0000 (11:50 +0800)]
Remove OpenCV stuffs from the memory data layer and io utils

10 years agoAdd lint rule for caffe data layer setup
Kai Li [Sun, 31 Aug 2014 09:47:58 +0000 (17:47 +0800)]
Add lint rule for caffe data layer setup

10 years agoFix conflict between nvcc and boost for cmake
Kai Li [Sat, 30 Aug 2014 16:16:31 +0000 (00:16 +0800)]
Fix conflict between nvcc and boost for cmake

10 years agoMove the rest duplicate codes of the data layers into their base class
Kai Li [Fri, 29 Aug 2014 08:37:59 +0000 (16:37 +0800)]
Move the rest duplicate codes of the data layers into their base class

10 years agoTest adding images w/o resizing to the memory data layer
Kai Li [Fri, 29 Aug 2014 08:36:04 +0000 (16:36 +0800)]
Test adding images w/o resizing to the memory data layer

10 years agoMove transform param one level up in the proto to reduce redundancy
Kai Li [Fri, 29 Aug 2014 03:45:37 +0000 (11:45 +0800)]
Move transform param one level up in the proto to reduce redundancy

10 years agoRemove pthread which has been replaced with boost thread
Kai Li [Fri, 29 Aug 2014 02:12:57 +0000 (10:12 +0800)]
Remove pthread which has been replaced with boost thread

10 years agoAdd transformer to the memory data layer
Kai Li [Thu, 28 Aug 2014 19:02:10 +0000 (03:02 +0800)]
Add transformer to the memory data layer

10 years agoImplement Forward_gpu in the base prefetching data layer
Kai Li [Thu, 28 Aug 2014 09:45:39 +0000 (17:45 +0800)]
Implement Forward_gpu in the base prefetching data layer

10 years agoThe BasePrefetchingDataLayer shouldn't join the thread
Kai Li [Thu, 28 Aug 2014 09:32:46 +0000 (17:32 +0800)]
The BasePrefetchingDataLayer shouldn't join the thread

10 years agoSimplify the WindowDataLayer using the base class
Kai Li [Thu, 28 Aug 2014 08:55:56 +0000 (16:55 +0800)]
Simplify the WindowDataLayer using the base class

10 years agoRemove duplicate codes from the ImageDataLayer
Kai Li [Thu, 28 Aug 2014 08:22:41 +0000 (16:22 +0800)]
Remove duplicate codes from the ImageDataLayer

10 years agoExtract common data layer functionalities out of the DataLayer
Kai Li [Thu, 28 Aug 2014 08:10:47 +0000 (16:10 +0800)]
Extract common data layer functionalities out of the DataLayer

10 years agoCreate base data layer and base prefetching data layer
Kai Li [Thu, 28 Aug 2014 07:29:25 +0000 (15:29 +0800)]
Create base data layer and base prefetching data layer

10 years agoMerge pull request #1025 from mohomran/minor_fix_to_mnist_solver_prototxt
Evan Shelhamer [Tue, 2 Sep 2014 15:38:07 +0000 (08:38 -0700)]
Merge pull request #1025 from mohomran/minor_fix_to_mnist_solver_prototxt

Snapshot naming issue in mnist_autoencoder_solver_adagrad.prototxt

10 years agofixed relative path and prefix for adagrad-optimised autoencoder snapshots
Mohamed Omran [Tue, 2 Sep 2014 11:01:13 +0000 (13:01 +0200)]
fixed relative path and prefix for adagrad-optimised autoencoder snapshots

10 years agoMerge pull request #1023 from longjon/unbreak-pycaffe
Evan Shelhamer [Tue, 2 Sep 2014 03:05:57 +0000 (20:05 -0700)]
Merge pull request #1023 from longjon/unbreak-pycaffe

Fix pycaffe break due to dropping {blob,layer}.name

10 years ago[pycaffe] use _blob_names, _layer_names instead of removed .name
Jonathan L Long [Tue, 2 Sep 2014 02:32:24 +0000 (19:32 -0700)]
[pycaffe] use _blob_names, _layer_names instead of removed .name

10 years ago[pycaffe] expose Net.blob_names and Net.layer_names
Jonathan L Long [Tue, 2 Sep 2014 02:26:02 +0000 (19:26 -0700)]
[pycaffe] expose Net.blob_names and Net.layer_names

10 years ago[pycaffe] add converter for vector<string> used by _*_names
Jonathan L Long [Tue, 2 Sep 2014 02:36:49 +0000 (19:36 -0700)]
[pycaffe] add converter for vector<string> used by _*_names

10 years agoadd CUDA 6.5 error CUBLAS_STATUS_LICENSE_ERROR to cublasGetErrorString
Jeff Donahue [Mon, 1 Sep 2014 23:45:20 +0000 (16:45 -0700)]
add CUDA 6.5 error CUBLAS_STATUS_LICENSE_ERROR to cublasGetErrorString
enum

10 years agorevert tools/train_net.cpp to previous, depecated version
Jeff Donahue [Mon, 1 Sep 2014 21:51:59 +0000 (14:51 -0700)]
revert tools/train_net.cpp to previous, depecated version

10 years agoMerge branch 'qipeng-solvers' into dev
Jeff Donahue [Mon, 1 Sep 2014 21:13:56 +0000 (14:13 -0700)]
Merge branch 'qipeng-solvers' into dev

10 years agomake MNIST autoencoder solvers start from base_lr 0.01 and step (much
Jeff Donahue [Mon, 1 Sep 2014 19:35:03 +0000 (12:35 -0700)]
make MNIST autoencoder solvers start from base_lr 0.01 and step (much
better performance) and terminate at iter 65K

10 years agomake adagrad/nesterov train scripts follow new "run-from-root"
Jeff Donahue [Mon, 1 Sep 2014 19:32:54 +0000 (12:32 -0700)]
make adagrad/nesterov train scripts follow new "run-from-root"
convention

10 years agoAdd "test-on-train" stage to test accuracy on the training data; correct
Jeff Donahue [Mon, 1 Sep 2014 19:04:31 +0000 (12:04 -0700)]
Add "test-on-train" stage to test accuracy on the training data; correct
test_iter (should be 100 instead of 50)

10 years agomnist_autoencoder: always compute both cross-entropy loss and L2
Jeff Donahue [Mon, 1 Sep 2014 18:39:39 +0000 (11:39 -0700)]
mnist_autoencoder: always compute both cross-entropy loss and L2
(euclidean) error

10 years agohot fix for warning
qipeng [Tue, 26 Aug 2014 19:21:06 +0000 (12:21 -0700)]
hot fix for warning

10 years agolint
qipeng [Tue, 26 Aug 2014 19:02:20 +0000 (12:02 -0700)]
lint

10 years agoRe-added solver switch into the new caffe main excutable; fixed AdaGrad MNIST example
qipeng [Tue, 26 Aug 2014 19:01:26 +0000 (12:01 -0700)]
Re-added solver switch into the new caffe main excutable; fixed AdaGrad MNIST example

10 years agolint
qipeng [Tue, 26 Aug 2014 06:17:51 +0000 (23:17 -0700)]
lint

10 years agoAdded sanity check for AdaGradSolver; added MNIST examples for solvers
qipeng [Tue, 26 Aug 2014 06:02:56 +0000 (23:02 -0700)]
Added sanity check for AdaGradSolver; added MNIST examples for solvers

10 years agoMerge Test{SGD,AdaGrad,Nesterov}Solver; they become subclasses of
Jeff Donahue [Fri, 22 Aug 2014 18:51:16 +0000 (11:51 -0700)]
Merge Test{SGD,AdaGrad,Nesterov}Solver; they become subclasses of
TestGradientBasedSolver

10 years agocleanup caffe.proto
Jeff Donahue [Fri, 22 Aug 2014 18:51:29 +0000 (11:51 -0700)]
cleanup caffe.proto

10 years agoadded unit test for solvers and fixed solver bugs
qipeng [Thu, 21 Aug 2014 03:07:53 +0000 (20:07 -0700)]
added unit test for solvers and fixed solver bugs

10 years agoproto conflit, lint, and math_functions (compiler complaint)
qipeng [Wed, 20 Aug 2014 22:19:30 +0000 (15:19 -0700)]
proto conflit, lint, and math_functions (compiler complaint)

10 years agofixes after rebase
qipeng [Wed, 30 Jul 2014 02:56:36 +0000 (19:56 -0700)]
fixes after rebase

10 years agoAddressed Yangqing's comments
qipeng [Wed, 30 Jul 2014 02:46:37 +0000 (19:46 -0700)]
Addressed Yangqing's comments

10 years agofixed caffe.proto after a mistaken rebase
qipeng [Tue, 29 Jul 2014 17:06:32 +0000 (10:06 -0700)]
fixed caffe.proto after a mistaken rebase

10 years agoAdded L1 regularization support for the weights
qipeng [Thu, 24 Jul 2014 20:09:28 +0000 (13:09 -0700)]
Added L1 regularization support for the weights

10 years agobugfixes for AdaGrad
qipeng [Wed, 23 Jul 2014 17:25:44 +0000 (10:25 -0700)]
bugfixes for AdaGrad

10 years agoimproved numerical stability for AdaGrad
qipeng [Wed, 23 Jul 2014 04:17:19 +0000 (21:17 -0700)]
improved numerical stability for AdaGrad

10 years agofixed solver constructor in train_net.cpp
qipeng [Mon, 21 Jul 2014 17:47:55 +0000 (10:47 -0700)]
fixed solver constructor in train_net.cpp

10 years agoconverted pointers to shared_ptr
qipeng [Sun, 20 Jul 2014 15:57:33 +0000 (08:57 -0700)]
converted pointers to shared_ptr

10 years agorestored vituals in solver.hpp
qipeng [Sun, 20 Jul 2014 00:33:35 +0000 (17:33 -0700)]
restored vituals in solver.hpp

10 years agoSolver switching support & implementation of Nesterov's accelerated gradient and...
qipeng [Sat, 19 Jul 2014 21:14:21 +0000 (14:14 -0700)]
Solver switching support & implementation of Nesterov's accelerated gradient and AdaGrad

10 years agouse LMDB in mnist autoencoder examples
Jeff Donahue [Mon, 1 Sep 2014 18:32:12 +0000 (11:32 -0700)]
use LMDB in mnist autoencoder examples

10 years agomake no GPU error in CPU-only mode a little clearer
Evan Shelhamer [Mon, 1 Sep 2014 06:28:28 +0000 (23:28 -0700)]
make no GPU error in CPU-only mode a little clearer

10 years agoinclude comment on CPU mode fine-tuning for Flickr example
Evan Shelhamer [Mon, 1 Sep 2014 06:25:31 +0000 (23:25 -0700)]
include comment on CPU mode fine-tuning for Flickr example

10 years agoMerge pull request #1014 from longjon/cleaner-pycaffe
Evan Shelhamer [Mon, 1 Sep 2014 06:22:21 +0000 (23:22 -0700)]
Merge pull request #1014 from longjon/cleaner-pycaffe

Clean up pycaffe core

10 years agoMerge pull request #1008 from mohomran/mnist_with_lmdb
Evan Shelhamer [Sun, 31 Aug 2014 18:02:45 +0000 (11:02 -0700)]
Merge pull request #1008 from mohomran/mnist_with_lmdb

convert MNIST demo to lmdb, fixes

10 years agoMerge pull request #1015 from mohomran/fixing_parse_log_script
Evan Shelhamer [Sun, 31 Aug 2014 18:00:48 +0000 (11:00 -0700)]
Merge pull request #1015 from mohomran/fixing_parse_log_script

fix for tools/extra/parse_log.sh

10 years agoparse_log.sh adapted to new training log format + fixed typos and updated description
Mohamed Omran [Sun, 31 Aug 2014 13:00:09 +0000 (15:00 +0200)]
parse_log.sh adapted to new training log format + fixed typos and updated description

10 years agoupdated lenet_train_test.prototxt + minor correction to create_mnist.sh
Mohamed Omran [Sun, 31 Aug 2014 12:27:33 +0000 (14:27 +0200)]
updated lenet_train_test.prototxt + minor correction to create_mnist.sh

10 years agominor changes to variable names and error messages + set default backed in convert_mn...
Mohamed Omran [Fri, 29 Aug 2014 15:16:41 +0000 (17:16 +0200)]
minor changes to variable names and error messages + set default backed in convert_mnist_data.cpp to lmdb

10 years agodata now written to backend in batches
Mohamed Omran [Fri, 29 Aug 2014 15:04:25 +0000 (17:04 +0200)]
data now written to backend in batches

10 years agomnist demo now works with lmdb and leveldb (set parameter in create_mnist.sh), switch...
Mohamed Omran [Fri, 29 Aug 2014 14:04:24 +0000 (16:04 +0200)]
mnist demo now works with lmdb and leveldb (set parameter in create_mnist.sh), switched around includes

10 years ago[pycaffe] expose Blob.Reshape
Jonathan L Long [Sun, 31 Aug 2014 08:14:58 +0000 (01:14 -0700)]
[pycaffe] expose Blob.Reshape

While this function should never be called in the current interface, it
will be necessary to have when embedding Python.

10 years ago[pycaffe] remove name property from PyBlob and PyLayer
Jonathan L Long [Sun, 31 Aug 2014 07:45:06 +0000 (00:45 -0700)]
[pycaffe] remove name property from PyBlob and PyLayer

Attaching names to pycaffe Blobs and Layers isn't necessary anymore,
since they are now accessed by name. This change brings PyBlob and
PyLayer closer to their C++ counterparts, and makes it easier to pass
around blobs without knowing their names, e.g., when embedding Python
in a layer.

10 years ago[pycaffe] declare the _caffe module init function
Jonathan L Long [Sun, 31 Aug 2014 07:36:26 +0000 (00:36 -0700)]
[pycaffe] declare the _caffe module init function

10 years ago[pycaffe] split _caffe into source and header files
Jonathan L Long [Sun, 31 Aug 2014 07:26:12 +0000 (00:26 -0700)]
[pycaffe] split _caffe into source and header files

This brings pycaffe more in line with usual C++ style, and will allow
the wrapper classes to be used within caffe when embedding Python.

10 years ago[pycaffe] make PyBlob a template over Dtype
Jonathan L Long [Sun, 31 Aug 2014 07:07:43 +0000 (00:07 -0700)]
[pycaffe] make PyBlob a template over Dtype

We will not actually support double nets in Python with this commit, but
this change will be necessary to create layers that embed Python.

10 years ago[pycaffe] make PyNet a class, not a struct
Jonathan L Long [Sun, 31 Aug 2014 07:01:20 +0000 (00:01 -0700)]
[pycaffe] make PyNet a class, not a struct