Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Sync upstream caffe #20

Open
wants to merge 545 commits into
base: master
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
545 commits
Select commit Hold shift + click to select a range
cdd2d0e
Fix: docs/yum_install.md glog broken link
Shamanoid Oct 5, 2016
a2f9e9b
Merge pull request #4812 from wk910930/fix-typo
shelhamer Oct 5, 2016
4ba654f
Merge pull request #4813 from Shamanoid/patch-1
shelhamer Oct 5, 2016
553a645
pytest fix: Files created with NamedTemporary files cannot be opened …
willyd Oct 14, 2016
95a436c
Fix: made load_hd5 check blob dims by default.
BlGene Aug 25, 2016
197d11a
sigmoid cross-entropy loss: add GPU forward for full GPU mode
shelhamer Oct 27, 2016
6491504
Merge pull request #4908 from shelhamer/sigce-gpu
shelhamer Oct 27, 2016
f59dc97
add the missing star in comment
nihui Nov 1, 2016
0d20df5
corrected typo in accuracy_layer.hpp: MaxTopBlos -> MaxTopBlobs
karas84 Nov 1, 2016
74f1fc2
Merge pull request #4937 from nihui/patch-1
jeffdonahue Nov 1, 2016
efe1732
Merge pull request #4941 from karas84/caffe-pr
jeffdonahue Nov 1, 2016
3b443ea
Add Github issue template to curb misuse.
williford Oct 28, 2016
29cdee7
Merge pull request #4914 from williford/issue_template
shelhamer Nov 3, 2016
20feab5
Put quotes around titles in YAML front matter.
williford Nov 11, 2016
eb4ba30
Merge pull request #4971 from williford/docs-installation-add-quotes
shelhamer Nov 12, 2016
aaf7b6b
support solver resumes in parse_log.py
davidbrai Nov 14, 2016
c6ab965
sigmoid cross-entropy loss: ignore selected targets by `ignore_label`
shelhamer Nov 15, 2016
6486a7b
docs: Guillaume Dumont is the Windows maintainer
shelhamer Nov 16, 2016
abcb973
docs: include AWS AMI pointer
shelhamer Nov 16, 2016
4a158a8
Merge pull request #4993 from shelhamer/bitfusion-ami
shelhamer Nov 16, 2016
3d62e3c
sigmoid cross-entropy loss: normalize loss by different schemes
shelhamer Nov 17, 2016
28c135c
Merge pull request #4986 from shelhamer/sigce-ignore
shelhamer Nov 17, 2016
2cf9dd3
Add missing spaces besides equal signs in batch_norm_layer.cpp
chenzeyuczy Nov 18, 2016
32c2ce6
Merge pull request #4979 from davidbrai/allow-parse_log-to-skip-lines
shelhamer Nov 18, 2016
473f143
Merge pull request #4703 from cypof/avoid_missing_returns
shelhamer Nov 19, 2016
e52451d
solver: check and set type to reconcile class and proto
shelhamer Nov 21, 2016
48e73c7
Checks inside Xcode for latest OSX SDK (#4840)
Zylphrex Nov 21, 2016
db6cf0a
Fix Python net drawing script
ngaloppo Nov 21, 2016
a60c4a4
Merge pull request #4998 from chenzeyuczy/master
jeffdonahue Nov 22, 2016
a6c6533
Merge pull request #5009 from shelhamer/solver-type-check
shelhamer Nov 22, 2016
2e59864
fix error link
kingofctrl Nov 24, 2016
b644a87
Revert "solver: check and set type to reconcile class and proto"
shelhamer Nov 25, 2016
db66432
fix many typos by using codespell
cdluminate Nov 27, 2016
e6b079d
Merge pull request #5035 from CDLuminate/fix-spelling
shelhamer Nov 27, 2016
fa7fda7
Make lint happy (> 80 characters)
cdluminate Nov 28, 2016
cd681ec
Add the missing period
cdluminate Nov 28, 2016
24d2f67
Merge pull request #5038 from CDLuminate/fix-spelling
jeffdonahue Nov 28, 2016
8cd5c3d
Add Pascal to all cuda architectures
Queuecumber Dec 2, 2016
5a201dd
Merge pull request #5057 from Queuecumber/cuda-pascal-cmake
shelhamer Dec 2, 2016
de3a12f
fix wrongly used marker hash
younghwanoh Dec 7, 2016
57a5bbd
check leveldb iterator status for snappy format.
Dec 16, 2016
b55fe84
docs: add debian installation guide
cdluminate Dec 21, 2016
2fac0d6
Use mkl_malloc when use mkl
tsocha Dec 8, 2016
775f5b0
Fixed a typo
yrevar Dec 22, 2016
3edfd7e
Merge pull request #5112 from yrevar/patch-1
jeffdonahue Dec 22, 2016
5693f31
Join path using "os.path.join" instead of "+"
yrevar Dec 23, 2016
1fd8bd0
Typos in test_inner_product_layer.cpp
ftokarev Dec 30, 2016
f731bc4
Merge pull request #5139 from ftokarev/patches
jeffdonahue Dec 30, 2016
4f0eb52
Fix parse_log.py and parse_log.sh for negative time duration if datet…
xiaojiedeng Dec 31, 2016
bae0607
Overhaul layer catalogue documentation.
williford Jan 1, 2017
fb52c7c
Add Debian codenames and make link.
williford Dec 21, 2016
5c437b1
docs: update debian installation guide. Thanks to @lukeyeager for com…
cdluminate Jan 4, 2017
369a1f4
docs: add some tables to debian install guide and misc update
cdluminate Jan 5, 2017
2317fa1
Logging from python, e.g. for lower log level on multi-GPU workers
cypof Nov 22, 2016
3ba2054
Switched multi-GPU to NCCL
cypof Jan 6, 2017
e21b420
Python Multi-GPU
cypof Nov 23, 2016
0d27efc
Python layers should build on multiprocess & solver_cnt; enable with …
mglaeser Dec 22, 2016
5f28eb1
Using default from proto for prefetch
cypof Jan 4, 2017
365ac88
Merge pull request #4338 from raffienficiaud/master
shelhamer Jan 7, 2017
8e63bb6
minor typo
stoneyang Jan 12, 2017
91c15e8
Python 2/3 compatible download_model_binary.py
willyd Jan 13, 2017
8314363
Merge pull request #5111 from CDLuminate/add-debian-install-guide
shelhamer Jan 13, 2017
0a91794
Merge pull request #5075 from tsocha/master
cypof Jan 14, 2017
a19357a
fixing upgrade_proto for BatchNorm layer: be more conservative leave …
Jan 15, 2017
536851c
Merge pull request #4563 from cypof/nccl
shelhamer Jan 17, 2017
8065c18
Merge pull request #5181 from willyd/download-model-py3
shelhamer Jan 17, 2017
6cbc72a
Merge pull request #5148 from caffe-help/caffe-docs-PR1
shelhamer Jan 18, 2017
ceb25c8
Fix various documentation typos (#4172)
abrowne2 Jan 18, 2017
510249b
Merge pull request #5176 from stoneyang/minor
shelhamer Jan 18, 2017
85e6b67
Merge pull request #5142 from lvchaxj/master
shelhamer Jan 18, 2017
9a980fd
Merge pull request #5121 from yrevar/patch-2
shelhamer Jan 18, 2017
63ca2f9
Merge pull request #5098 from yaronli/master
shelhamer Jan 18, 2017
381c022
Merge pull request #5025 from hmybmny/master
shelhamer Jan 18, 2017
e744056
remove redundant operations in Crop layer (#5138)
xmyqsh Jan 18, 2017
0c8bc25
Merge pull request #4842 from willyd/pytest-fix
shelhamer Jan 18, 2017
9b9f6d0
[build] remove trailing backslash on comment
longjon Jan 19, 2017
ff3158a
ignore generated includes for docs
shelhamer Jan 19, 2017
9ab6709
copyright spans 2014-2017
shelhamer Jan 19, 2017
c86433a
Merge pull request #3365 from BonsaiAI/wrap-declarations-in-switch
shelhamer Jan 19, 2017
bc0d680
Merge pull request #5184 from shaibagon/fix_batch_norm_param_upgrade
shelhamer Jan 20, 2017
5f84ad7
Merge pull request #5198 from longjon/config-backslash
shelhamer Jan 20, 2017
4056f79
Docker refresh: simplified & update to 16.04, cuda8, cudnn5, nccl
cypof Jan 18, 2017
b8fa34d
Merge pull request #5153 from cypof/docker
cypof Jan 20, 2017
1354403
cmake: bump soversion to rc4
cdluminate Jan 21, 2017
3a0b6c6
Update a comment in caffe.proto
ftokarev Jan 21, 2017
3babd4e
Merge pull request #5207 from CDLuminate/cmake-bump-soversion-to-rc4
shelhamer Jan 22, 2017
e0cd852
Restore can be invoked on rank > 0
cypof Jan 23, 2017
51f2986
Merge pull request #5215 from cypof/fix_restore
cypof Jan 23, 2017
29f0cdb
parse_log.py was not using --verbose argument
kts Jan 24, 2017
b015b73
Merge pull request #5220 from kts/bugfix
shelhamer Jan 24, 2017
6bf10af
Fix broken links in layer documentation, minor fixes.
williford Jan 20, 2017
746a77e
Merge pull request #5227 from williford/caffe-git-pr2
shelhamer Jan 26, 2017
7b5731c
Remove sdk version from veclib searching path.
solrex Jan 30, 2017
cd89d4b
docs: update install_apt_debian guide
cdluminate Feb 1, 2017
39f28e4
Merge pull request #5242 from CDLuminate/doc-update-debian-guide
shelhamer Feb 1, 2017
734702b
Document switch to explicit flags for docker: cpu / gpu.
cypof Feb 8, 2017
9c201e1
make: bump version to rc4
shelhamer Feb 9, 2017
15dfcc1
Add Pascal CUDA architectures to Makefile.config.example
crowsonkb Feb 9, 2017
d8f7953
Merge pull request #5272 from crowsonkb/master
shelhamer Feb 10, 2017
929135b
Merge pull request #5210 from ftokarev/patches
shelhamer Feb 16, 2017
28ffe9c
Merge pull request #5236 from solrex/veclib-fix
shelhamer Feb 16, 2017
c510e33
Merge pull request #4347 from nitnelave/python/layer_dict
shelhamer Feb 17, 2017
c5fad40
Merge pull request #3893 from rscohn2/patch-1
shelhamer Feb 17, 2017
8a7a61a
Merge pull request #5074 from garion9013/master
shelhamer Feb 17, 2017
4b65348
Merge pull request #5010 from ngaloppo/fix_drawnet
shelhamer Feb 17, 2017
25dfdfb
Merge pull request #4721 from kashefy/fix_matlab_demo_typos
shelhamer Feb 17, 2017
475d619
Merge pull request #4737 from rokm/matcaffe-individual-destruct
shelhamer Feb 17, 2017
9bb1b9c
Merge pull request #4609 from intelfx/BVLC-work-buildsystem
shelhamer Feb 17, 2017
23fca12
version bump: rc5
shelhamer Feb 17, 2017
16467ff
Merge pull request #5296 from shelhamer/rc5
shelhamer Feb 21, 2017
85ab610
fix broken link to hinge loss
shelhamer Feb 27, 2017
fe9e58d
Remove not used variable in base_conv_layer.cpp
zhuyuanhao Mar 1, 2017
e687a71
Merge pull request #4630 from BlGene/load_hdf5_fix
shelhamer Mar 8, 2017
4529f12
Removed some 'warning: extra ‘;’ [-Wpedantic]'
gineshidalgo99 Mar 10, 2017
1d3e6e4
Solver_add_nccl accepts any kind of Solver
jfolz Mar 13, 2017
317d162
Merge pull request #5393 from jfolz/master
cypof Mar 13, 2017
93993a3
Init test net on all GPUs, allows parallel inference
cypof Mar 14, 2017
802d90f
Added python 3 compatibility to cpp_lint.py
willyd Mar 17, 2017
accd188
sane h5df file type check for weights
BlGene Mar 8, 2017
367a6e2
Merge pull request #5372 from BlGene/hdf5-load-fix
erictzeng Mar 21, 2017
11930f1
Clarify batch norm parameter documentation.
williford Mar 22, 2017
5c8e354
[caffe][build] added Atlas lapack Library name atllapack
BlGene Mar 22, 2017
1e02d62
[caffe][build] added ABS_TEST_DATA_DIR var.
BlGene Mar 22, 2017
793bd96
Merge pull request #5434 from williford/batchnorm_doc2
shelhamer Mar 23, 2017
8602a23
Expose share_weights to python to allow running test nets
cypof Mar 27, 2017
850ffd8
Remove missed legacy parallel code
cypof Mar 27, 2017
9bd80b2
Fix typo in test_caffe_main.cpp: defice -> device
YuduoWu Mar 29, 2017
a32114e
Fixed memory leaks in cudnn conv and relu
willcrichton Mar 31, 2017
a2601ed
Revert "Fix Python net drawing script"
lukeyeager Mar 31, 2017
0096fe3
Add support for cuDNN v6
flx42 Mar 31, 2017
179dafd
Add test for caffe.draw.draw_net()
lukeyeager Mar 31, 2017
5ad9e53
Merge pull request #5478 from flx42/cudnn6-support
shelhamer Mar 31, 2017
c0597b1
Merge pull request #5477 from lukeyeager/bvlc/test-draw-net
shelhamer Mar 31, 2017
41e34c9
Minor fix for net drawing script
nitheeshas Apr 4, 2017
31bfe8f
Add main() for draw_net unittest, fix import errors
nitheeshas Apr 4, 2017
5f1ca84
Add example and small blurb about sigmoid layer.
williford Apr 5, 2017
e9f7941
Merge pull request #5487 from nitheeshas/fix-draw-net
shelhamer Apr 6, 2017
daf0139
Merge pull request #5491 from caffe-help/sigmoid-doc-1
shelhamer Apr 6, 2017
af2a626
Merge pull request #5467 from yuduowu/master
shelhamer Apr 6, 2017
7b9f2ac
Merge pull request #5380 from gineshidalgo99/pull-request-supressed-s…
shelhamer Apr 7, 2017
ce7193c
Removed repeated import Layer, get_solver
willyd Apr 8, 2017
b2a95fa
Log shape dimensions for eltwise layer shape mismatch
brunobowden Apr 8, 2017
88b10af
Merge pull request #5503 from brunobowden/shape_check_eq
shelhamer Apr 12, 2017
90eff9b
Merge pull request #5455 from cypof/remove_shared_parallel
cypof Apr 12, 2017
41a7d21
Merge pull request #5408 from cypof/multi_infer
cypof Apr 12, 2017
31a18c2
Merge pull request #5474 from willcrichton/master
cypof Apr 12, 2017
51728d1
Fix log parsing #5422
cypof Apr 12, 2017
bac59be
Allow using env vars for glog init from python
cypof Apr 12, 2017
35a7b87
fixes pycaffe forward() and backward() behavior for nets whose layer …
Noiredd Apr 12, 2017
31e634b
Merge pull request #5420 from willyd/py3-lint
shelhamer Apr 12, 2017
02f1c58
Merge pull request #5515 from cypof/py_glog_env
shelhamer Apr 12, 2017
6ba1486
Merge pull request #5506 from willyd/pycaffe-fix
shelhamer Apr 13, 2017
3a98796
remove redundant check in LSTMUnitLayer
Apr 13, 2017
c43eb0b
Merge pull request #5437 from BlGene/test-path-fix
shelhamer Apr 13, 2017
fecee6c
Merge pull request #5337 from zhuyuanhao/master
shelhamer Apr 13, 2017
9687062
Bump boost version to 1.55 in CMake build
willyd Apr 13, 2017
0c9cc62
Added support for python 3 and NCCL
willyd Mar 12, 2017
e98023a
Add GPU sqrt functions
Dec 29, 2016
e93fcd2
GPU BatchNormLayer: replace powx with mul and sqrt
Dec 29, 2016
ab33988
Add CPU sqrt functions
jeffdonahue Apr 13, 2017
1c15d94
CPU BatchNormLayer: replace powx with sqr and sqrt
jeffdonahue Apr 13, 2017
c560658
Merge pull request #5136 from pfollmann/fix_batchnorm_layer
jeffdonahue Apr 13, 2017
5e6ffd7
Merge pull request #4182 from ajschumacher/handle_dtypes
shelhamer Apr 13, 2017
3d5bed0
fix: add non-MKL sqrt (should have been included in ab33988)
jeffdonahue Apr 13, 2017
f522d9c
Merge pull request #5521 from kkhoot/fix_lstm_unit_check
jeffdonahue Apr 13, 2017
2ec19b6
deprecate WindowData layer type
shelhamer Apr 13, 2017
a54b943
Merge pull request #5527 from willyd/nccl-py3
cypof Apr 13, 2017
e7163f6
Updated Travis boost dependencies
willyd Apr 14, 2017
ea2d786
Merge pull request #5526 from willyd/boost-1_55
shelhamer Apr 14, 2017
8bc82c6
[examples] switch cifar-10 back to proto instead of h5 serialization
shelhamer Apr 14, 2017
c425f23
Merge pull request #4576 from CDLuminate/add-bash-completion
shelhamer Apr 14, 2017
950e003
Merge pull request #4237 from CDLuminate/cmake-using-gnuinstalldirs
shelhamer Apr 14, 2017
aa29eba
Explicit std::string to bp::object conversion
willyd Apr 14, 2017
e6b2ba5
Merge pull request #5519 from Noiredd/master
shelhamer Apr 14, 2017
c7abdae
Merge pull request #3825 from jasjuang/master
shelhamer Apr 14, 2017
c19c960
Test for python forward and backward with start and end layer.
cdoersch Aug 25, 2015
861a12d
Merge pull request #4566 from CDLuminate/fix-more-float-comparison-issue
shelhamer Apr 14, 2017
1836b1b
Merge pull request #5514 from cypof/fix_parse_log
shelhamer Apr 14, 2017
4519443
[docs] added apt command to install OpenBLAS (#4718)
jgyllinsky Apr 14, 2017
2cbc1bb
Merge pull request #3855 from shaibagon/upgrade_infogain
shelhamer Apr 14, 2017
c11e782
Merge pull request #3913 from antran89/master
shelhamer Apr 14, 2017
1df3a25
Merge pull request #3153 from jeffdonahue/netspec-type-check
shelhamer Apr 14, 2017
6137b24
Merge pull request #3411 from ghost/patch-3
shelhamer Apr 14, 2017
08bded8
Merge pull request #3410 from ghost/patch-2
shelhamer Apr 14, 2017
68215cc
Merge pull request #2612 from ih4cku/master
shelhamer Apr 14, 2017
8007349
fix lint errors that snuck in by #4566
shelhamer Apr 14, 2017
4db619a
Docker update to cuDNN 6
cypof Apr 14, 2017
44da39f
BVLC -> BAIR
shelhamer Apr 14, 2017
3562698
drop performance + hardware page and switch to sheet
shelhamer Apr 14, 2017
0f5bfc3
favor notebook examples as more clear and popular
shelhamer Apr 14, 2017
2158bbb
model zoo: point out wiki link immediately, explain manual editing
shelhamer Apr 14, 2017
414b74c
add missing names to BAIR roster
shelhamer Apr 14, 2017
e90a6a6
retire caffe-dev and caffe-coldpress
shelhamer Apr 14, 2017
8985818
track publications by google scholar and not the wiki
shelhamer Apr 14, 2017
8b8f2dd
link to new full-day crash course
shelhamer Apr 14, 2017
49761d3
Caffe 1.0
shelhamer Apr 14, 2017
28780e9
Merge pull request #5529 from shelhamer/deprecate-window
shelhamer Apr 14, 2017
2e33792
Merge pull request #5531 from shelhamer/py-test-layer-top-names
shelhamer Apr 14, 2017
946c9b8
Merge pull request #5537 from shelhamer/docs-grooming
shelhamer Apr 14, 2017
a7a7d30
Merge pull request #5536 from cypof/docker_cudnn6
shelhamer Apr 15, 2017
69cf20a
Merge pull request #5539 from shelhamer/caffe-1.0
shelhamer Apr 15, 2017
eeebdab
Merge pull request #5530 from willyd/nccl-py3
cypof Apr 15, 2017
33f8612
Rewrite crop cuda kernel
erictzeng Apr 19, 2017
cd1696d
Fix crop layer lint errors
erictzeng Apr 19, 2017
ec35395
Handling destruction of empty Net objects
May 4, 2017
c293d9d
Merge pull request #5588 from ShaggO/matlab-fix-delete
shelhamer May 4, 2017
7d3f8a7
Merge pull request #5548 from erictzeng/crop
shelhamer May 4, 2017
b7e2b99
Downgrade boost requirement from 1.55 to 1.54
lukeyeager May 12, 2017
25391bf
Merge pull request #5617 from lukeyeager/boost-1.54
shelhamer May 12, 2017
30a2ab7
cmake: rename libproto.a -> libcaffeproto.a
cdluminate May 15, 2017
83814da
docs/debian guide: update compiler combination table
cdluminate May 15, 2017
27ffbbf
Merge pull request #5625 from CDLuminate/docs-update
shelhamer May 15, 2017
91b0928
Merge pull request #5624 from CDLuminate/cmake-rename-static-proto-li…
shelhamer May 15, 2017
264cf19
List branches in readme
cypof Jun 13, 2017
f16b5f2
Merge pull request #5687 from BVLC/readme_list_branches
cypof Jun 14, 2017
4efdf7e
Update README.md
cypof Jun 20, 2017
c209c92
Fixed bug where make distribute duplicates python files in distribute…
Jun 21, 2017
4a2f2a0
docs: update apt installation guide for Debian and Ubuntu
cdluminate Jul 8, 2017
eedf7c1
docs: add Ubuntu package tracker link in Ubuntu guide
cdluminate Jul 8, 2017
3d7cfc4
Update lrn.md
Lydorn Jul 10, 2017
af9d6bc
update sklearn calls to use latest API
Jul 12, 2017
4b98f06
update deprecated pandas call
Jul 12, 2017
751a570
[DOC][FIX] fix web demo install instruction link
jqueguiner Aug 1, 2017
f10e24b
Merge pull request #5770 from skylarjhdownes/master
shelhamer Aug 3, 2017
516c135
Merge pull request #5760 from Lydorn/patch-1
shelhamer Aug 3, 2017
7ad1342
Merge pull request #5753 from CDLuminate/docs-update-deb-ub
shelhamer Aug 3, 2017
315641b
Fix hardcode xcode path
ek9852 Aug 4, 2017
86d451e
Merge pull request #5824 from ek9852/master
shelhamer Aug 6, 2017
1de4ceb
Update README.md
cypof Aug 17, 2017
1f1326d
Update link to google style guide.
cijianzy Aug 21, 2017
3dad332
modified division operator for compatibility of python 3
wasnot Sep 6, 2017
ea455eb
Merge pull request #5900 from wasnot/fix/py3-division-compat
longjon Sep 6, 2017
8bdc87f
Expose GPU pointers to Python
tnarihi Mar 11, 2015
4705a37
Packages needed by Ubuntu 16.04 also
williford Sep 15, 2017
effcdb0
Merge pull request #5904 from longjon/gpu-ptr
shelhamer Sep 19, 2017
888597e
Fixed bilinear filler, added tests
Noiredd Oct 2, 2017
ef2eb4b
Merge pull request #5713 from Noiredd/filler
shelhamer Oct 3, 2017
fac7434
[docs] fix link to `AbsVal` layer
shelhamer Oct 3, 2017
c7ae0e2
Fix default mode warning in io.resize_image
developius Oct 10, 2017
62e0c85
upgrading Accuracy layer: (1) efficient CPU implementation O(L) for t…
Aug 10, 2017
243cd89
Add absolute tolerance to test_net.py to prevent random Travis fails
Noiredd Oct 11, 2017
691febc
Merge pull request #5925 from BVLC/williford-install-ubuntu-16.04-pat…
shelhamer Oct 11, 2017
79ddda7
infogain loss: fix bottom blobs description
anmikh Oct 17, 2017
b4ffad8
Merge pull request #5973 from Noiredd/pytest
Noiredd Oct 19, 2017
d03f812
Merge pull request #5866 from cijianzy/update_link_to_google_style_guide
Noiredd Nov 2, 2017
388a410
Merge pull request #5969 from developius/fix-default-mode-warning
Noiredd Nov 2, 2017
2c733c2
Merge pull request #5704 from ArneSuppe/dupDistDirFix
Noiredd Nov 2, 2017
b536e15
Merge pull request #5813 from jqueguiner/patch-1
Noiredd Nov 2, 2017
0dff5d3
Upgrade to latest caffe
asnt Nov 4, 2017
05d01fb
Adapt proto for deepcut following caffe sync
asnt Nov 7, 2017
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
19 changes: 19 additions & 0 deletions .github/ISSUE_TEMPLATE.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,19 @@
Please use the [caffe-users list](https://groups.google.com/forum/#!forum/caffe-users) for usage, installation, or modeling questions, or other requests for help.
_Do not post such requests to Issues._ Doing so interferes with the development of Caffe.

Please read the [guidelines for contributing](https://github.com/BVLC/caffe/blob/master/CONTRIBUTING.md) before submitting this issue.

### Issue summary


### Steps to reproduce

If you are having difficulty building Caffe or training a model, please ask the caffe-users mailing list. If you are reporting a build error that seems to be due to a bug in Caffe, please attach your build configuration (either Makefile.config or CMakeCache.txt) and the output of the make (or cmake) command.

### Your system configuration
Operating system:
Compiler:
CUDA version (if applicable):
CUDNN version (if applicable):
BLAS:
Python or MATLAB version (for pycaffe and matcaffe respectively):
4 changes: 4 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -47,6 +47,9 @@
# PyCharm files
.idea

# Visual Studio Code files
.vscode

# OSX dir files
.DS_Store

Expand Down Expand Up @@ -81,6 +84,7 @@ cmake_build

# Generated documentation
docs/_site
docs/_includes
docs/gathered
_site
doxygen
Expand Down
58 changes: 35 additions & 23 deletions .travis.yml
Original file line number Diff line number Diff line change
@@ -1,40 +1,52 @@
# Use a build matrix to do two builds in parallel:
# one using CMake, and one using make.
dist: trusty
sudo: required

language: cpp
compiler: gcc

env:
global:
- NUM_THREADS=4
matrix:
- WITH_CUDA=false WITH_CMAKE=false WITH_IO=true
- WITH_CUDA=false WITH_CMAKE=true WITH_IO=true PYTHON_VERSION=3
- WITH_CUDA=true WITH_CMAKE=false WITH_IO=true
- WITH_CUDA=true WITH_CMAKE=true WITH_IO=true
- WITH_CUDA=false WITH_CMAKE=false WITH_IO=false
- WITH_CUDA=false WITH_CMAKE=true WITH_IO=false PYTHON_VERSION=3
# Use a build matrix to test many builds in parallel
# envvar defaults:
# WITH_CMAKE: false
# WITH_PYTHON3: false
# WITH_IO: true
# WITH_CUDA: false
# WITH_CUDNN: false
- BUILD_NAME="default-make"
# - BUILD_NAME="python3-make" WITH_PYTHON3=true
- BUILD_NAME="no-io-make" WITH_IO=false
- BUILD_NAME="cuda-make" WITH_CUDA=true
- BUILD_NAME="cudnn-make" WITH_CUDA=true WITH_CUDNN=true

language: cpp
- BUILD_NAME="default-cmake" WITH_CMAKE=true
- BUILD_NAME="python3-cmake" WITH_CMAKE=true WITH_PYTHON3=true
- BUILD_NAME="no-io-cmake" WITH_CMAKE=true WITH_IO=false
- BUILD_NAME="cuda-cmake" WITH_CMAKE=true WITH_CUDA=true
- BUILD_NAME="cudnn-cmake" WITH_CMAKE=true WITH_CUDA=true WITH_CUDNN=true

# Cache Ubuntu apt packages.
cache:
apt: true
directories:
- /home/travis/miniconda
- /home/travis/miniconda2
- /home/travis/miniconda3

compiler: gcc
- ~/protobuf3

before_install:
- export NUM_THREADS=4
- export SCRIPTS=./scripts/travis
- export CONDA_DIR="/home/travis/miniconda$PYTHON_VERSION"
- source ./scripts/travis/defaults.sh

install:
- sudo -E $SCRIPTS/travis_install.sh
- sudo -E ./scripts/travis/install-deps.sh
- ./scripts/travis/setup-venv.sh ~/venv
- source ~/venv/bin/activate
- ./scripts/travis/install-python-deps.sh

before_script:
- export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib:/usr/local/cuda/lib64:$CONDA_DIR/lib
- export PATH=$CONDA_DIR/bin:$PATH
- if ! $WITH_CMAKE; then $SCRIPTS/travis_setup_makefile_config.sh; fi
- ./scripts/travis/configure.sh

script: $SCRIPTS/travis_build_and_test.sh
script:
- ./scripts/travis/build.sh
- ./scripts/travis/test.sh

notifications:
# Emails are sent to the committer's git-configured email address by default,
Expand Down
44 changes: 38 additions & 6 deletions CMakeLists.txt
Original file line number Diff line number Diff line change
Expand Up @@ -10,14 +10,15 @@ endif()
project(Caffe C CXX)

# ---[ Caffe version
set(CAFFE_TARGET_VERSION "1.0.0-rc3")
set(CAFFE_TARGET_SOVERSION "1.0.0-rc3")
set(CAFFE_TARGET_VERSION "1.0.0" CACHE STRING "Caffe logical version")
set(CAFFE_TARGET_SOVERSION "1.0.0" CACHE STRING "Caffe soname version")
add_definitions(-DCAFFE_VERSION=${CAFFE_TARGET_VERSION})

# ---[ Using cmake scripts and modules
list(APPEND CMAKE_MODULE_PATH ${PROJECT_SOURCE_DIR}/cmake/Modules)

include(ExternalProject)
include(GNUInstallDirs)

include(cmake/Utils.cmake)
include(cmake/Targets.cmake)
Expand All @@ -28,6 +29,7 @@ include(cmake/ConfigGen.cmake)
# ---[ Options
caffe_option(CPU_ONLY "Build Caffe without CUDA support" OFF) # TODO: rename to USE_CUDA
caffe_option(USE_CUDNN "Build Caffe with cuDNN library support" ON IF NOT CPU_ONLY)
caffe_option(USE_NCCL "Build Caffe with NCCL library support" OFF)
caffe_option(BUILD_SHARED_LIBS "Build shared libraries" ON)
caffe_option(BUILD_python "Build Python wrapper" ON)
set(python_version "2" CACHE STRING "Specify which Python version to use")
Expand All @@ -38,6 +40,7 @@ caffe_option(USE_OPENCV "Build with OpenCV support" ON)
caffe_option(USE_LEVELDB "Build with levelDB" ON)
caffe_option(USE_LMDB "Build with lmdb" ON)
caffe_option(ALLOW_LMDB_NOLOCK "Allow MDB_NOLOCK when reading LMDB files (only if necessary)" OFF)
caffe_option(USE_OPENMP "Link with OpenMP (when your BLAS wants OpenMP and you get linker errors)" OFF)

# ---[ Dependencies
include(cmake/Dependencies.cmake)
Expand All @@ -47,13 +50,13 @@ if(UNIX OR APPLE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fPIC -Wall")
endif()

caffe_set_caffe_link()

if(USE_libstdcpp)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -stdlib=libstdc++")
message("-- Warning: forcing libstdc++ (controlled by USE_libstdcpp option in cmake)")
endif()

add_definitions(-DGTEST_USE_OWN_TR1_TUPLE)

# ---[ Warnings
caffe_warnings_disable(CMAKE_CXX_FLAGS -Wno-sign-compare -Wno-uninitialized)

Expand All @@ -62,8 +65,26 @@ configure_file(cmake/Templates/caffe_config.h.in "${PROJECT_BINARY_DIR}/caffe_co

# ---[ Includes
set(Caffe_INCLUDE_DIR ${PROJECT_SOURCE_DIR}/include)
include_directories(${Caffe_INCLUDE_DIR} ${PROJECT_BINARY_DIR})
include_directories(BEFORE src) # This is needed for gtest.
set(Caffe_SRC_DIR ${PROJECT_SOURCE_DIR}/src)
include_directories(${PROJECT_BINARY_DIR})

# ---[ Includes & defines for CUDA

# cuda_compile() does not have per-call dependencies or include pathes
# (cuda_compile() has per-call flags, but we set them here too for clarity)
#
# list(REMOVE_ITEM ...) invocations remove PRIVATE and PUBLIC keywords from collected definitions and include pathes
if(HAVE_CUDA)
# pass include pathes to cuda_include_directories()
set(Caffe_ALL_INCLUDE_DIRS ${Caffe_INCLUDE_DIRS})
list(REMOVE_ITEM Caffe_ALL_INCLUDE_DIRS PRIVATE PUBLIC)
cuda_include_directories(${Caffe_INCLUDE_DIR} ${Caffe_SRC_DIR} ${Caffe_ALL_INCLUDE_DIRS})

# add definitions to nvcc flags directly
set(Caffe_ALL_DEFINITIONS ${Caffe_DEFINITIONS})
list(REMOVE_ITEM Caffe_ALL_DEFINITIONS PRIVATE PUBLIC)
list(APPEND CUDA_NVCC_FLAGS ${Caffe_ALL_DEFINITIONS})
endif()

# ---[ Subdirectories
add_subdirectory(src/gtest)
Expand All @@ -83,8 +104,19 @@ if(BUILD_python)
add_dependencies(pytest pycaffe)
endif()

# ---[ uninstall target
configure_file(
${CMAKE_CURRENT_SOURCE_DIR}/cmake/Uninstall.cmake.in
${CMAKE_CURRENT_BINARY_DIR}/cmake/Uninstall.cmake
IMMEDIATE @ONLY)

add_custom_target(uninstall
COMMAND ${CMAKE_COMMAND} -P
${CMAKE_CURRENT_BINARY_DIR}/cmake/Uninstall.cmake)

# ---[ Configuration summary
caffe_print_configuration_summary()

# ---[ Export configs generation
caffe_generate_export_configs()

2 changes: 1 addition & 1 deletion CONTRIBUTORS.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
# Contributors

Caffe is developed by a core set of BVLC members and the open-source community.
Caffe is developed by a core set of BAIR members and the open-source community.

We thank all of our [contributors](https://github.com/BVLC/caffe/graphs/contributors)!

Expand Down
4 changes: 2 additions & 2 deletions LICENSE
Original file line number Diff line number Diff line change
@@ -1,11 +1,11 @@
COPYRIGHT

All contributions by the University of California:
Copyright (c) 2014, 2015, The Regents of the University of California (Regents)
Copyright (c) 2014-2017 The Regents of the University of California (Regents)
All rights reserved.

All other contributions:
Copyright (c) 2014, 2015, the respective contributors
Copyright (c) 2014-2017, the respective contributors
All rights reserved.

Caffe uses a shared copyright model: each contributor holds copyright over
Expand Down
42 changes: 30 additions & 12 deletions Makefile
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,7 @@ LIB_BUILD_DIR := $(BUILD_DIR)/lib
STATIC_NAME := $(LIB_BUILD_DIR)/lib$(LIBRARY_NAME).a
DYNAMIC_VERSION_MAJOR := 1
DYNAMIC_VERSION_MINOR := 0
DYNAMIC_VERSION_REVISION := 0-rc3
DYNAMIC_VERSION_REVISION := 0
DYNAMIC_NAME_SHORT := lib$(LIBRARY_NAME).so
#DYNAMIC_SONAME_SHORT := $(DYNAMIC_NAME_SHORT).$(DYNAMIC_VERSION_MAJOR)
DYNAMIC_VERSIONED_NAME_SHORT := $(DYNAMIC_NAME_SHORT).$(DYNAMIC_VERSION_MAJOR).$(DYNAMIC_VERSION_MINOR).$(DYNAMIC_VERSION_REVISION)
Expand Down Expand Up @@ -192,12 +192,12 @@ ifeq ($(USE_LMDB), 1)
LIBRARIES += lmdb
endif
ifeq ($(USE_OPENCV), 1)
LIBRARIES += opencv_core opencv_highgui opencv_imgproc
LIBRARIES += opencv_core opencv_highgui opencv_imgproc

ifeq ($(OPENCV_VERSION), 3)
LIBRARIES += opencv_imgcodecs
endif

endif
PYTHON_LIBRARIES ?= boost_python python2.7
WARNINGS := -Wall -Wno-sign-compare
Expand Down Expand Up @@ -248,6 +248,8 @@ ifeq ($(UNAME), Linux)
LINUX := 1
else ifeq ($(UNAME), Darwin)
OSX := 1
OSX_MAJOR_VERSION := $(shell sw_vers -productVersion | cut -f 1 -d .)
OSX_MINOR_VERSION := $(shell sw_vers -productVersion | cut -f 2 -d .)
endif

# Linux
Expand All @@ -270,22 +272,29 @@ endif
ifeq ($(OSX), 1)
CXX := /usr/bin/clang++
ifneq ($(CPU_ONLY), 1)
CUDA_VERSION := $(shell $(CUDA_DIR)/bin/nvcc -V | grep -o 'release \d' | grep -o '\d')
CUDA_VERSION := $(shell $(CUDA_DIR)/bin/nvcc -V | grep -o 'release [0-9.]*' | tr -d '[a-z ]')
ifeq ($(shell echo | awk '{exit $(CUDA_VERSION) < 7.0;}'), 1)
CXXFLAGS += -stdlib=libstdc++
LINKFLAGS += -stdlib=libstdc++
endif
# clang throws this warning for cuda headers
WARNINGS += -Wno-unneeded-internal-declaration
# 10.11 strips DYLD_* env vars so link CUDA (rpath is available on 10.5+)
OSX_10_OR_LATER := $(shell [ $(OSX_MAJOR_VERSION) -ge 10 ] && echo true)
OSX_10_5_OR_LATER := $(shell [ $(OSX_MINOR_VERSION) -ge 5 ] && echo true)
ifeq ($(OSX_10_OR_LATER),true)
ifeq ($(OSX_10_5_OR_LATER),true)
LDFLAGS += -Wl,-rpath,$(CUDA_LIB_DIR)
endif
endif
endif
# gtest needs to use its own tuple to not conflict with clang
COMMON_FLAGS += -DGTEST_USE_OWN_TR1_TUPLE=1
# boost::thread is called boost_thread-mt to mark multithreading on OS X
LIBRARIES += boost_thread-mt
# we need to explicitly ask for the rpath to be obeyed
DYNAMIC_FLAGS := -install_name @rpath/libcaffe.so
ORIGIN := @loader_path
VERSIONFLAGS += -Wl,-install_name,$(DYNAMIC_VERSIONED_NAME_SHORT) -Wl,-rpath,$(ORIGIN)/../../build/lib
VERSIONFLAGS += -Wl,-install_name,@rpath/$(DYNAMIC_VERSIONED_NAME_SHORT) -Wl,-rpath,$(ORIGIN)/../../build/lib
else
ORIGIN := \$$ORIGIN
endif
Expand Down Expand Up @@ -319,6 +328,12 @@ ifeq ($(USE_CUDNN), 1)
COMMON_FLAGS += -DUSE_CUDNN
endif

# NCCL acceleration configuration
ifeq ($(USE_NCCL), 1)
LIBRARIES += nccl
COMMON_FLAGS += -DUSE_NCCL
endif

# configure IO libraries
ifeq ($(USE_OPENCV), 1)
COMMON_FLAGS += -DUSE_OPENCV
Expand Down Expand Up @@ -355,9 +370,9 @@ ifeq ($(BLAS), mkl)
# MKL
LIBRARIES += mkl_rt
COMMON_FLAGS += -DUSE_MKL
MKL_DIR ?= /opt/intel/mkl
BLAS_INCLUDE ?= $(MKL_DIR)/include
BLAS_LIB ?= $(MKL_DIR)/lib $(MKL_DIR)/lib/intel64
MKLROOT ?= /opt/intel/mkl
BLAS_INCLUDE ?= $(MKLROOT)/include
BLAS_LIB ?= $(MKLROOT)/lib $(MKLROOT)/lib/intel64
else ifeq ($(BLAS), open)
# OpenBLAS
LIBRARIES += openblas
Expand All @@ -373,8 +388,11 @@ else
LIBRARIES += cblas
# 10.10 has accelerate while 10.9 has veclib
XCODE_CLT_VER := $(shell pkgutil --pkg-info=com.apple.pkg.CLTools_Executables | grep 'version' | sed 's/[^0-9]*\([0-9]\).*/\1/')
XCODE_CLT_GEQ_7 := $(shell [ $(XCODE_CLT_VER) -gt 6 ] && echo 1)
XCODE_CLT_GEQ_6 := $(shell [ $(XCODE_CLT_VER) -gt 5 ] && echo 1)
ifeq ($(XCODE_CLT_GEQ_6), 1)
ifeq ($(XCODE_CLT_GEQ_7), 1)
BLAS_INCLUDE ?= /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/$(shell ls /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/ | sort | tail -1)/System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/Headers
else ifeq ($(XCODE_CLT_GEQ_6), 1)
BLAS_INCLUDE ?= /System/Library/Frameworks/Accelerate.framework/Versions/Current/Frameworks/vecLib.framework/Headers/
LDFLAGS += -framework Accelerate
else
Expand Down Expand Up @@ -552,7 +570,7 @@ $(ALL_BUILD_DIRS): | $(BUILD_DIR_LINK)

$(DYNAMIC_NAME): $(OBJS) | $(LIB_BUILD_DIR)
@ echo LD -o $@
$(Q)$(CXX) -shared -o $@ $(OBJS) $(VERSIONFLAGS) $(LINKFLAGS) $(LDFLAGS) $(DYNAMIC_FLAGS)
$(Q)$(CXX) -shared -o $@ $(OBJS) $(VERSIONFLAGS) $(LINKFLAGS) $(LDFLAGS)
@ cd $(BUILD_DIR)/lib; rm -f $(DYNAMIC_NAME_SHORT); ln -s $(DYNAMIC_VERSIONED_NAME_SHORT) $(DYNAMIC_NAME_SHORT)

$(STATIC_NAME): $(OBJS) | $(LIB_BUILD_DIR)
Expand Down Expand Up @@ -676,6 +694,6 @@ $(DISTRIBUTE_DIR): all py | $(DISTRIBUTE_SUBDIRS)
install -m 644 $(DYNAMIC_NAME) $(DISTRIBUTE_DIR)/lib
cd $(DISTRIBUTE_DIR)/lib; rm -f $(DYNAMIC_NAME_SHORT); ln -s $(DYNAMIC_VERSIONED_NAME_SHORT) $(DYNAMIC_NAME_SHORT)
# add python - it's not the standard way, indeed...
cp -r python $(DISTRIBUTE_DIR)/python
cp -r python $(DISTRIBUTE_DIR)/

-include $(DEPS)
Loading