Граф коммитов

1495 Коммитов

Автор SHA1 Сообщение Дата
Sergey Karayev 9c61462cac back-merging [docs] changes and web demo [example] addition; updating
net_surgery example to new format

Conflicts:
	docs/getting_pretrained_models.md
	docs/index.md
2014-07-12 09:25:23 -07:00
Sergey Karayev dd292da248 Merge pull request #679 from sergeyk/master
[example] image classification web demo
2014-07-12 09:20:07 -07:00
Sergey Karayev dd546171ba [example] image classification web demo 2014-07-11 23:41:12 -07:00
Sergey Karayev e8e8292c36 Merge pull request #675 from sergeyk/master
Revamped docs system
2014-07-11 17:20:46 -07:00
Sergey Karayev 05d661f042 [docs] shelhamer's minor suggestions 2014-07-11 17:19:35 -07:00
Sergey Karayev 54c94190a9 [docs] updated instructions for contributing documentation 2014-07-11 14:00:56 -07:00
Sergey Karayev 0de282ce21 [docs] moved example md’s to examples/**/md’s and added script to gather
them for publication
2014-07-11 13:49:07 -07:00
Sergey Karayev 583b84ef65 [docs] cosmetic 2014-07-11 13:29:06 -07:00
Evan Shelhamer e07a2c0eea fix link for caffe_rcnn_imagenet_model 2014-07-11 06:09:41 +02:00
Evan Shelhamer 923b040db6 Back-merge documentation and fixes
* master:
  host materials on dl.caffe.berkeleyvision.org
  fix caffe acm-mm paper link
  [docs] readme
  [docs] got rid of redundant README, updated development instructions
  [docs] reworked index page, got rid of publications page
  point to reference performance from installation, add GTX 770
  acknowledge BVLC PI Trevor Darrell for advising Caffe
  add latest CUDA arch to fix invalid device function errors
  switch language to "related publications"
  add publication section to homepage
  fix caffe paper link -- still hasn't appeared on arxiv yet
  Added top-1 and top-5 accuracy for the caffe networks to docs
  add skeleton of the Caffe publications page
  Update docs on building boost on OSX for the python wrappers
  fix OSX 10.9 homebrew CXX doc
  caffe.Net preprocessing members belong to object, not class
  10.9 install doc formatting
2014-07-11 06:07:55 +02:00
Evan Shelhamer 7bd30fe6ba host materials on dl.caffe.berkeleyvision.org 2014-07-11 06:01:42 +02:00
Evan Shelhamer 7e2520da88 fix caffe acm-mm paper link 2014-07-10 14:45:44 +02:00
longjon efa7176577 Merge pull request #633 from kloudkl/cpu-only-memcpy
Support CPU only memcpy
2014-07-09 17:18:33 -07:00
Kai Li 904c2ce69b Replace cudaMemcpy with caffe_gpu_memcpy in SyncedMemory per @longjon 2014-07-10 08:03:22 +08:00
Kai Li ac0dd39252 Implement @Yangqing's solution to copy memory in the SyncedMemory 2014-07-10 08:03:22 +08:00
Kai Li 00433d8082 Switch to GPU mode when pointer is move to or from GPU in SyncedMemory 2014-07-10 08:03:22 +08:00
Kai Li 4d16ed5969 Check the GPU mode to decide which memcpy to use 2014-07-10 08:03:22 +08:00
Kai Li 3bd49bac72 Avoid using cudaMemcpy for memcpy when there is no GPU and CUDA driver 2014-07-10 08:03:22 +08:00
Jeff Donahue 1679815f67 makefile hotfix 2014-07-09 16:39:17 -07:00
Jeff Donahue e5f72b7910 Merge pull request #661 from jeffdonahue/fix-error-output
Fix bug introduced by warning logs (errors didn't print)
2014-07-09 15:54:35 -07:00
Jeff Donahue 37329cbc2c fix bug introduced by warning logs: errors didn't print because they
are logged to the warnings file
2014-07-09 15:47:21 -07:00
Sergey Karayev 9223cf4476 [docs] readme 2014-07-09 15:07:42 -07:00
Sergey Karayev a7e43ed28f [docs] got rid of redundant README, updated development instructions 2014-07-09 15:06:10 -07:00
Jeff Donahue d86735ed3b Makefile: cleanup lint/warn report logic (and make the two more consistent) 2014-07-09 11:42:19 -07:00
Sergey Karayev f0cc3f7ee1 [docs] reworked index page, got rid of publications page 2014-07-09 11:32:34 -07:00
Evan Shelhamer 368288ea3c Merge pull request #649 from jeffdonahue/warnlog
"make warn" to print compilation warnings; "make lint" runs incrementally
2014-07-09 08:47:36 +02:00
Jeff Donahue 933dc6e7d9 Output a lint report for every source file linted; use to lint
incrementally
2014-07-08 17:56:06 -07:00
Jeff Donahue ba8a2bb8c9 Dump compiler warnings to *.warnings.txt; use "make warn" to print them 2014-07-08 17:53:53 -07:00
Jeff Donahue dc21bc12a5 Merge pull request #647 from jeffdonahue/nocleanlinecount
Don't make clean when running linecount
2014-07-08 15:06:34 -07:00
Jeff Donahue e5be75459a Don't make clean when running linecount 2014-07-08 15:02:00 -07:00
Jeff Donahue d82a1ea6bc Merge pull request #614 from ronghanghu/rectangular_pooling
Rectangular pooling
2014-07-07 16:40:50 -07:00
Ronghang Hu 45e201d42d added gradient check for non-square pooling 2014-07-07 12:12:59 -07:00
Ronghang Hu e10977f8a5 fixed style errors 2014-07-07 11:32:58 -07:00
Evan Shelhamer 34d1e93732 Merge pull request #611 from shelhamer/makefile-config-cxx
Customize compiler setting in Makefile.config
2014-07-06 12:47:53 -07:00
Ronghang Hu f74979e7de add tests for rectangular pooling regions 2014-07-05 08:21:50 -07:00
Ronghang Hu 7b727b020d fixing pooling SetUp() to allow default values for stride and pad 2014-07-05 08:11:03 -07:00
Ronghang Hu ce88f9f966 Update pooling_layer.cu
Replace pad_, kernel_size_, stride_ with pad_h_, pad_w_, kernel_size_h_, kernel_size_w_, stride_h_, stride_w_ to support pooling on rectangle regions.
2014-07-03 21:05:23 -07:00
Ronghang Hu dd82d97190 Update pooling_layer.cpp
Replace pad_, kernel_size_, stride_ with pad_h_, pad_w_, kernel_size_h_, kernel_size_w_, stride_h_, stride_w_ to support pooling on rectangle regions.
2014-07-03 20:42:58 -07:00
Ronghang Hu 9905c20130 Update vision_layers.hpp
Replace pad_, kernel_size_, stride_ with pad_h_, pad_w_, kernel_size_h_, kernel_size_w_, stride_h_, stride_w_ to support pooling on rectangle regions.
2014-07-03 20:33:50 -07:00
Ronghang Hu 4e5ef959ab Update caffe.proto
Add pad_h, pad_w, kernel_size_h, kernel_size_w, stride_h, stride_w to support pooling on rectangle regions.
2014-07-03 20:30:25 -07:00
Evan Shelhamer 03a48d7d4e point to reference performance from installation, add GTX 770 2014-07-03 19:07:10 -07:00
Evan Shelhamer 18a4e2d1fc customize compiler setting in Makefile.config
in case of issues with default compilers or exotic platforms.
2014-07-03 18:38:30 -07:00
Evan Shelhamer e09e256284 Merge pull request #549 from jamt9000/fix-resize-crop-pil
Make resizing & cropping with PIL work
2014-07-03 17:29:14 -07:00
Evan Shelhamer aef3de7366 Merge pull request #555 from shelhamer/uva-memory
Switch to Unified Virtual Address memory copies
2014-07-03 17:27:37 -07:00
Jeff Donahue eb40cb460a Merge pull request #602 from kloudkl/layers-in-order
Layers in order
2014-07-03 17:25:51 -07:00
Evan Shelhamer 19e909ba98 fix casts (static for void*) 2014-07-03 17:20:31 -07:00
Evan Shelhamer 6461365565 reduce caffe_copy to instantiations, split off caffe_memcpy for void* 2014-07-03 17:20:30 -07:00
Evan Shelhamer af624faa55 replace all memset with caffe_set() / caffe_gpu_set()
...except for `SyncedMem` since it has no type.
2014-07-03 17:20:28 -07:00
Evan Shelhamer 1da30a5558 replace all memcpy by caffe_copy 2014-07-03 17:14:12 -07:00
Evan Shelhamer aaa9f24d36 do all caffe_copy() as UVA mem copy, and drop caffe_gpu_copy()
Do all memory copies by `cudaMemcpy` in UVA mode so that the same
`caffe_copy()` interface works for all transfers.

`cudaMemcpy()` is used in lieu of BLAS copies because they do not
understand UVA.

Drop the now unnecessary `caffe_gpu_copy()` since location of the
pointers is now irrelevant to the interface.
2014-07-03 17:14:12 -07:00