Commits

Breadcrumbs

History for

onparallel

User selector

All users

All time

Commit History

Commits on Dec 21, 2014

• Added credits for training bvlc models

  sguada

  committed
  f347355

  Copy full SHA for f347355

• Added bvlc_googlenet prototxt and weights

  sguada

  committed
  f15210c

  Copy full SHA for f15210c

Commits on Oct 17, 2014

• back-merge

  sergeyk

  committed
  1ed11d5

  Copy full SHA for 1ed11d5

• Fixing finetune_flickr_style model reported accuracy.

  sergeyk

  committed
  f2f0a7a

  Copy full SHA for f2f0a7a

Commits on Oct 4, 2014

• bundle pixel mean into CaffeNet as comments

  shelhamer

  committed
  3115b20

  Copy full SHA for 3115b20

Commits on Oct 3, 2014

• Fixed crop error and add test_data_transformer

  Show description for b07a54f
  sguada

  committed
  b07a54f

  Copy full SHA for b07a54f

• Fix OSX compilation for nvcc with opencv

  sguada

  committed
  8cd863e

  Copy full SHA for 8cd863e

• Added mean_value to specify mean channel substraction

  Show description for a9572b1
  sguada

  committed
  a9572b1

  Copy full SHA for a9572b1

Commits on Sep 29, 2014

• [examples] adding class names and deploy version of Flickr Style net

  sergeyk

  committed
  c68f9ce

  Copy full SHA for c68f9ce

Commits on Sep 22, 2014

• switch examples to lmdb (except for custom data loaders)

  shelhamer

  committed
  73f1de4

  Copy full SHA for 73f1de4

Commits on Sep 4, 2014

• fix fine-tuning example: paths, test acc., and total fine-tuning time

  shelhamer

  committed
  a857001

  Copy full SHA for a857001

• fix model download advice and prototxt name for fine-tuning

  shelhamer

  committed
  f7baf2b

  Copy full SHA for f7baf2b

• script to upload/update model info as gist

  sergeyk

  committed
  51c4e6e

  Copy full SHA for 51c4e6e

• flickr style fine-tuning model (separated from example read me)

  sergeyk

  committed
  c6827bf

  Copy full SHA for c6827bf

• Renaming CaffeNet model prototxts and unignoring models/*

  sergeyk

  committed
  a661001

  Copy full SHA for a661001

• proofread model zoo

  

  shelhamer

  authored and

  sergeyk

  committed
  39f7a4d

  Copy full SHA for 39f7a4d

• [models] adding zoo readme; caffenet, alexnet, and rcnn models in zoo format

  sergeyk

  committed
  e553573

  Copy full SHA for e553573

Commits on Mar 20, 2014

• file pascal finetuning prototxt examples and fix paths

  shelhamer

  committed
  32cea65

  Copy full SHA for 32cea65

• some cleanup

  

  rbgirshick

  authored and

  shelhamer

  committed
  f1a64b7

  Copy full SHA for f1a64b7

• fix paths

  

  rbgirshick

  authored and

  shelhamer

  committed
  458fa2b

  Copy full SHA for 458fa2b

• support for tightest square mode while finetuning

  

  rbgirshick

  authored and

  shelhamer

  committed
  1339297

  Copy full SHA for 1339297

• 10x learning rate for fine tuning makes a big difference

  

  rbgirshick

  authored and

  shelhamer

  committed
  5cb7c23

  Copy full SHA for 5cb7c23

• support for adding padding to windows in the window_data_layer

  

  rbgirshick

  authored and

  shelhamer

  committed
  17a0c16

  Copy full SHA for 17a0c16

• Code that was used to finetune with reasonable success

  

  rbgirshick

  authored and

  shelhamer

  committed
  bb0e72c

  Copy full SHA for bb0e72c

• adjustments to try to match the setup for fine tuning with cuda-convnet

  

  rbgirshick

  authored and

  shelhamer

  committed
  f40c30b

  Copy full SHA for f40c30b

• define pascal finetuning models

  

  rbgirshick

  authored and

  shelhamer

  committed
  f4a2c14

  Copy full SHA for f4a2c14

Commits on Mar 18, 2014

• drop models/ in favor of examples/

  Show description for d124b8f
  shelhamer

  committed
  d124b8f

  Copy full SHA for d124b8f

• minor fix to caffe model DL script

  sergeyk

  committed
  bc37a0f

  Copy full SHA for bc37a0f

• Add support for md5 checksum on OS X

  kloudkl

  committed
  41a126d

  Copy full SHA for 41a126d

• Avoid repeatedly downloading caffe reference imagenet model

  kloudkl

  committed
  9b9eb7a

  Copy full SHA for 9b9eb7a

Commits on Feb 26, 2014

• remove padding layers in imagenet definitions

  

  mavenlin

  authored and

  shelhamer

  committed
  2848aa1

  Copy full SHA for 2848aa1

• harmonize imagenet example, name caffe reference model CaffeNet

  shelhamer

  committed
  bb07c06

  Copy full SHA for bb07c06

• everything in its right place

  shelhamer

  committed
  04f61b3

  Copy full SHA for 04f61b3

• include model checksum, you never know these days

  shelhamer

  committed
  4fd2fa2

  Copy full SHA for 4fd2fa2

• fetch caffe_reference_imagenet_model

  shelhamer

  committed
  3230b5a

  Copy full SHA for 3230b5a