Proto commits in MVIG-SJTU/RMPE

These commits are when the Protocol Buffers files have changed: (only the last 100 relevant commits are shown)

2016-12-15

add shuffle option for data_heatmap layer

The documentation is generated from this commit.

2016-11-17

merge file filler @daerduoCarey

2016-11-02

update stn, add de_transform

2016-10-03

fix bug in prediction, add support for bbox scale

2016-10-01

update files

2016-09-26

add prediction heatmap layer

add detection heatmap layer

2016-09-11

final model, add UpsampleNearest filler, merge Eltwise Affine layer@ducha-aiki

2016-09-02

SHG runable

2016-08-31

add flip switch to heatmao_data_param

fix bug, add HG, pass simple test

2016-08-30

merge caffe-heatmap@tpfister

2016-08-29

Merge pull request #1 from weiliu89/ssd Merge ssd

2016-07-13

add demo for processing video file

2016-07-05

merge master and fix conflict

2016-06-28

add default value for rms_decay

2016-06-01

Add RecurrentLayer: an abstract superclass for other recurrent layer types

2016-05-05

add parameter layer for learning any bottom

2016-04-27

Fix an error in the example of ReshapeParameter. * this small mistake may confuse newer.

2016-04-09

rebase master

2016-04-05

Net: setting `propagate_down: true` forces backprop

2016-03-29

add webcam demo

2016-03-20

speed up nms and generate output for COCO

2016-03-06

add map_object_to_agnostic to enable learning object proposal

2016-03-05

Extend Crop to N-D, changed CropParameter.

Crop: fixes, tests and negative axis indexing.

add CropLayer: crop blob to another blob's dimensions with offsets configure offset(s) through proto definition.

2016-03-04

enable different choice of encoding the prior variance

make variance repeated in PriorBoxLayer

2016-02-29

merge master

2016-02-25

deprecate input fields and upgrade automagically

add InputLayer for Net input Create an input layer to replace oddball Net `input` fields.

add more normalization to MultiBoxLoss

2016-02-24

add logistic conf loss type

2016-02-21

add keep_top_k in DetectionOutputLayer

2016-02-20

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

2016-02-19

add sampler and related functions

2016-02-16

add support when evaluating on partial test data

2016-02-15

add code_type

2016-02-14

add variance and make max_size optional

add neg_overlap for selecting hard negatives

2016-02-11

do negative mining based on scores instead of overlap

2016-02-09

change type of size from int to float

add do_neg_mining in MultiBoxLossLayer

2016-02-08

add name_size_file in DetectionEvaluationLayer

2016-02-07

add SmoothL1LossLayer from Ross Girshick's Fast R-CNN

2016-02-06

add MaxIntegral ap_version to match VOC2012/ILSVRC AP

add difficult property for bbox annotation

2016-02-05

set lr_mult to 0 instead of using fix_scale in NormalizeLayer to not learn scale parameter

add num_classes in DetectionEvaluateLayer

2016-02-04

add SaveOutputParameter in DetectionOutputLayer

2016-02-03

add NormalizeLayer from fcn branch

2016-02-02

add change in proto for normalize option

add TestDetection

2016-02-01

add DetectionEvaluateLayer with test

add DetectionOutputLayer with test

2016-01-31

add ApplyNMS and GetConfidenceScores to bbox_util

2016-01-27

fix merge upstream conflict

2016-01-25

fix several bugs in MultiBoxLossLayer

2016-01-24

add PermuteLayer

2016-01-23

fix merge upstream conflict

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.

2016-01-22

Add ChannelwiseAffine for batch norm

put label_map_file in AnnotatedDataParameter

Merge pull request #3388 from mohomran/exponential_linear_units Exponential Linear Units

add MultiBoxLossLayer and bbox_util

2016-01-19

add PriorBoxLayer which generates priors from a layer

2016-01-15

add LabelMap and tools for create DB to store AnnotatedDatum

2016-01-14

Add AnnotatedDataLayer

2015-12-28

add short description of dilation to caffe.proto

add support for 2D dilated convolution

2015-12-04

ELU layer with basic tests

2015-11-22

Better normalization options for SoftmaxWithLoss layer.

2015-10-23

Merge pull request #3229 from cdoersch/batchnorm2 Yet another batch normalization PR

2015-10-22

Cleanup batch norm layer, include global stats computation

2015-10-17

Change solver type to string and provide solver registry

2015-10-16

Merge pull request #3160 from shelhamer/cudnnV3 Basic cuDNN v3 support

Initial cuDNN v3 support

2015-09-25

Add argmax_param axis

2015-09-19

Im2col and Convolution layers support N spatial axes

caffe.proto: generalize ConvolutionParameter to N spatial axes

2015-08-26

Add TileLayer

2015-08-25

Merge pull request #2032 from jeffdonahue/embed-layer Embed layer for lookup table of one hot encodings

2015-08-14

Adam solver 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.

2015-08-13

Merge pull request #2903 from ronghanghu/multi_gpu Multi-GPU Data Parallelism

2015-08-12

Data Layers Parallel for Multi-GPU Allow data layers (and also PythonLayer when used as data layer) to be shared among worker solver's training net, and also test net for future-proof if one wants to do Multi-GPU testing. Data layers are locked during forward to ensure sequential forward.

2015-08-10

Implement AdaDelta; add test cases; add mnist examples

2015-08-09

Add DataReader for parallel training with one DB session - Make sure each solver accesses a different subset of the data - Sequential reading of DB for performance - Prefetch a configurable amount of data to host memory - Distribute data to solvers in round-robin way for determinism

Implement RMSProp Solver Implement RMSProp solver and cleaned up to adjust to new solver interface that uses accumulated gradients and refactored regularization.

2015-08-08

Add EmbedLayer for inner products with sparse input (one-hot vectors), with unit tests

temporarily switch the snapshot_format default back to BINARYPROTO out of anticipation for user issues due to issue #2885, which causes Caffe to crash when it attempts to snapshot nets with duplicate layer names

2015-08-07

Snapshot model weights/solver state to HDF5 files. Summary of changes: - HDF5 helper functions were moved into a separate file util/hdf5.cpp - hdf5_save_nd_dataset now saves n-d blobs, can save diffs instead of data - Minor fix for memory leak in HDF5 functions (delete instead of delete[]) - Extra methods have been added to both Net/Solver enabling snapshotting and restoring from HDF5 files - snapshot_format was added to SolverParameters, with possible values HDF5 or BINARYPROTO (default HDF5) - kMaxBlobAxes was reduced to 32 to match the limitations of HDF5

add double_data, double_diff to BlobProto for weights/snapshots saved when using Dtype == double

PythonLayer takes parameters by string

2015-08-04

ImageData layer default batch size of 1, and check for zero batch size

2015-06-03

Add ReductionLayer to reduce any number of "tail" axes to a scalar value Currently implements operations SUM, MEAN, ASUM (sum of absolute values), and SUMSQ (sum of squares)

FlattenLayer gets a FlattenParameter with an axis, end_axis

Add LogLayer

2015-05-30

Merge pull request #1977 from shelhamer/accum-grad Decouple the computational batch size and minibatch size by accumulating gradients

Merge pull request #2410 from sguada/datum_transform Datum transform