platform/upstream/caffeonacl - Domain: Machine Learning / ML Framework; Licenses: BSD-2-Clause;

Age	Commit message (Collapse)	Author	Files	Lines
2015-09-17	Separate IO dependencies	Tea	1	-0/+12
	OpenCV, LMDB, LevelDB and Snappy are made optional via switches (USE_OPENCV, USE_LMDB, USE_LEVELDB) available for Make and CMake builds. Since Snappy is a LevelDB dependency, its use is determined by USE_LEVELDB. HDF5 is left bundled because it is used for serializing weights and solverstates.
2015-08-20	Use input_shape instead of input_dim in examples	Luke Yeager	1	-4/+6

2015-08-14	Adam solver	PatWie	2	-0/+29
	This commit implements the Adam solver by Kingma et. al for CPU and GPU. All solver parameters are defined in the caffe.proto. This also adds an example for the MNIST dataset.
2015-08-10	Clean up and modernize AdaDelta code; add learning rate support; add ↵	Matthias Plappert	2	-0/+4
	additional test cases
2015-08-10	Updated AdaDelta for modern Caffe; reduced iterations on multi-iter tests	Kevin Bache	1	-1/+1

2015-08-10	Implement AdaDelta; add test cases; add mnist examples	Mohamed Omran	3	-0/+43

2015-08-08	Implement RMSProp Solver	Eren Golge	2	-0/+30
	Implement RMSProp solver and cleaned up to adjust to new solver interface that uses accumulated gradients and refactored regularization.
2015-08-08	add [] to "delete pixels".	Tian Zhi	1	-1/+1
	see https://isocpp.org/wiki/faq/freestore-mgmt#delete-array-built-ins.
2015-06-30	[examples] sequence and revise notebooks	Evan Shelhamer	1	-0/+24
	- combine classification + filter visualization - order by classification, learning LeNet, brewing logreg, and fine-tuning to flickr style - improve flow of content in classification + filter visualization - include solver needed for learning LeNet - edit notebook descriptions for site catalogue
2015-06-15	typo: "a fixed steps" to "at fixed steps"	Aaron Schumacher	1	-1/+1
	fixing in the correct place as per @shelhamer's advice from #2602
2015-05-14	Update docs for ND blobs (#1970) and layer type is a string (#1694)	Jeff Donahue	1	-25/+25

2015-02-05	Upgrade existing nets using upgrade_net_proto_text tool	Jeff Donahue	3	-179/+250
	Restore comments afterwards
2015-01-25	[example] lenet early stopping -> mnist examples	Evan Shelhamer	1	-0/+28

2014-10-17	back-merge	Sergey Karayev	1	-9/+8

2014-10-16	Fix init current_step	Sergio	1	-6/+5

2014-10-15	Adapt lenet_multistep_solver.prototxt to current solvers	Sergio	2	-0/+33

2014-10-09	[example] re-title LeNet / MNIST heading too	Evan Shelhamer	1	-1/+1

2014-10-09	[example] add LeNet to MNIST title, fix paths to be from root	Evan Shelhamer	1	-8/+7

2014-09-21	switch examples to lmdb (except for custom data loaders)	Evan Shelhamer	3	-9/+9

2014-09-18	[docs] lenet grooming	Evan Shelhamer	1	-8/+11

2014-09-18	Updated mnist/readme.md file with additional information.	Christos Nikolaou	1	-13/+16

2014-09-17	Update readme.md files of cifar10 and mnist examples. Fixed broken links.	Christos Nikolaou	1	-3/+17

2014-09-07	fix transform_param in mnist_autoencoder.prototxt	Jeff Donahue	1	-3/+3

2014-09-04	[docs] default setting for layout	Sergey Karayev	1	-1/+0

2014-09-03	[example] convert mnist name fix (crashes xcode compiler)	qipeng	1	-1/+1

2014-09-03	[example] drop stale mentions of glog env var	Evan Shelhamer	1	-1/+1

2014-09-03	Inline latest lenet_solver.prototxt	Daniel Golden	1	-4/+4

2014-09-03	Correct reference to lenet_train_test.prototxt	Daniel Golden	1	-1/+1
	Not lenet.prototxt
2014-09-03	Point to local file, not github file	Daniel Golden	1	-1/+1
	This makes the readme file more consistent
2014-09-03	Update paths	Daniel Golden	1	-3/+3

2014-09-03	Move transform param one level up in the proto to reduce redundancy	Kai Li	3	-21/+21

2014-09-02	fixed relative path and prefix for adagrad-optimised autoencoder snapshots	Mohamed Omran	1	-1/+1

2014-09-01	make MNIST autoencoder solvers start from base_lr 0.01 and step (much	Jeff Donahue	3	-10/+14
	better performance) and terminate at iter 65K
2014-09-01	make adagrad/nesterov train scripts follow new "run-from-root"	Jeff Donahue	2	-4/+4
	convention
2014-09-01	Add "test-on-train" stage to test accuracy on the training data; correct	Jeff Donahue	4	-10/+39
	test_iter (should be 100 instead of 50)
2014-09-01	mnist_autoencoder: always compute both cross-entropy loss and L2	Jeff Donahue	1	-4/+4
	(euclidean) error
2014-09-01	Re-added solver switch into the new caffe main excutable; fixed AdaGrad ↵	qipeng	2	-2/+1
	MNIST example
2014-09-01	Added sanity check for AdaGradSolver; added MNIST examples for solvers	qipeng	4	-0/+38

2014-09-01	use LMDB in mnist autoencoder examples	Jeff Donahue	1	-2/+4

2014-08-31	updated lenet_train_test.prototxt + minor correction to create_mnist.sh	Mohamed Omran	2	-4/+6

2014-08-31	minor changes to variable names and error messages + set default backed in ↵	Mohamed Omran	1	-19/+26
	convert_mnist_data.cpp to lmdb
2014-08-31	data now written to backend in batches	Mohamed Omran	2	-15/+38

2014-08-31	mnist demo now works with lmdb and leveldb (set parameter in ↵	Mohamed Omran	3	-31/+101
	create_mnist.sh), switched around includes
2014-08-30	back-merge	Evan Shelhamer	1	-2/+2
	[docs] Several documentation fixes Explain how to fix homebrew to allow updates after editing formulae
2014-08-30	[docs] Several documentation fixes	Ozan Çağlayan	1	-2/+2
	- Fix a typo and a path in MNIST tutorial - Rework instructions about Fedora installation
2014-08-28	set examples paths relative to root	Evan Shelhamer	9	-27/+27

2014-08-21	upgrade model definitions for transformation params	Evan Shelhamer	3	-7/+21

2014-08-13	Add ACCURACY layer and softmax_error output to lenet_consolidated_solver	Jeff Donahue	1	-1/+0
	example.
2014-08-11	[docs] sorting of examples. if doesn’t work for you, update jekyll.	Sergey Karayev	1	-0/+1

2014-08-11	[docs] sorting of examples. if doesn’t work for you, update jekyll.	Sergey Karayev	1	-0/+1