Skip to content
Projects
Groups
Snippets
Help
Loading...
Sign in
Toggle navigation
D
dlib
Project
Project
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
钟尚武
dlib
Commits
ae1b1357
Commit
ae1b1357
authored
Jun 25, 2016
by
Davis King
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
updated docs
parent
15f4081c
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
47 additions
and
46 deletions
+47
-46
release_notes.xml
docs/docs/release_notes.xml
+47
-46
No files found.
docs/docs/release_notes.xml
View file @
ae1b1357
...
@@ -12,60 +12,61 @@
...
@@ -12,60 +12,61 @@
<current>
<current>
New Features:
New Features:
- Added find_optimal_parameters()
- A deep learning toolkit using CPU and/or GPU hardware. Some major elements
- Added elastic_net
of this are:
- Added random_color_transform and disturb_colors().
- Clean and fully documented C++11 API
- Added a constructor for seeding rand with a time_t.
- Clean tutorials: see dnn_introduction_ex.cpp and dnn_introduction2_ex.cpp
- Added apply_random_color_offset()
- Uses cuDNN v5.0
- Made load_image() support GIF files.
- Multi-GPU support
- Added subm_clipped()
- Automatic learning rate adjustment
- load_mnist_dataset()
- A pretrained 1000 class Imagenet classifier (see dnn_imagenet_ex.cpp)
- Added an option to solve the L2-loss version of the SVM objective function for svm_c_linear_dcd_trainer.
- Optimization Tools
- Added the option to use the elastic net regularizer to the OCA solver.
- Added find_optimal_parameters()
- Added solve_qp_box_constrained()
- Added elastic_net class
- Added unserialize.
- Added the option to use the elastic net regularizer to the OCA solver.
- MATLAB binding stuff
- Added an option to solve the L2-loss version of the SVM objective function to svm_c_linear_dcd_trainer.
- link to MATLAB's intel MKL when used on linux
- Added solve_qp_box_constrained()
- struct support, more arguments (20 now instead of 10),
- Image Processing
- in place operation. Made column major matrices directly wrap matlab
- Added random_color_transform, disturb_colors(), and apply_random_color_offset().
- load_image() now supports loading GIF files.
- Many improvements to the MATLAB binding API
- Automatically link to MATLAB's Intel MKL when used on linux.
- struct support
- mex functions can have up to 20 arguments instead of 10.
- In place operation. Made column major matrices directly wrap MATLAB
matrix objects when used inside mex files. This way, if you use
matrix objects when used inside mex files. This way, if you use
matrix_colmajor or fmatrix_colmajor in a mex file it will not do any
matrix_colmajor or fmatrix_colmajor in a mex file it will not do any
unnecessary copying or transposing.
unnecessary copying or transposing.
- catch ctrl+c presses in MATLAB console.
- Catch ctrl+c presses in MATLAB console. Allowing early termination of mex functions.
- DLIB_ASSERTS won't kill the matlab process, just throw an exception
- When used inside mex files, DLIB_ASSERTS won't kill the MATLAB process,
- Made cerr print in matlab as a red warning message.
just throw an exception.
- Made cerr print in MATLAB as a red warning message.
- load_mnist_dataset()
- C++11 only tools
- Added a constructor for seeding rand with a time_t.
- Added log1pexp()
- Added subm_clipped()
- Added running_gradient
- Added unserialize.
- deep learning tools
- Added running_gradient
- dnn_trainer
- cuDNN v4.0
- auto step size adjust and stopping condition.
- CUDA/tensor stuff
- gpu_data, tensor, alias tensors
Non-Backwards Compatible Changes:
Non-Backwards Compatible Changes:
- Everything in dlib/matlab/call_matlab.h is now in the dlib namespace.
- Everything in dlib/matlab/call_matlab.h is now in the dlib namespace.
- DLIB_TEST() and DLIB_TEST_MSG() macros now require you to terminate them with ;
- DLIB_TEST() and DLIB_TEST_MSG() macros now require you to terminate them with
a
;
Bug fixes:
Bug fixes:
- Fixed bug in 10 argument version of call_matlab() and also cleaned up a few
- Fixed bug in 10 argument version of call_matlab() and also cleaned up a few
minor things.
minor things.
- setup.py and
cm
ake scripts work in a few more contexts.
- setup.py and
CM
ake scripts work in a few more contexts.
-
Compiler errors in visual studio 2015
-
Fixed compiler errors in visual studio 2015.
- Fixed a bug in gaussian_blur() that caused messed up outputs when big
sigma values were used on some pixel types.
- Fixed a bug in gaussian_blur() that caused messed up outputs when big
- Fixed minor bugs in join_rows() and join_cols(). They didn't work when one of the matrices was empty
.
sigma values were used on some pixel types
.
- Fixed minor bugs in join_rows() and join_cols(). They didn't work when one
of the matrices was empty.
Other:
- Made cmake scripts uniformly require cmake version 2.8.4.
Other:
-
faster fHOG feature extraction / face detection
-
Made CMake scripts uniformly require CMake version 2.8.4.
-
C++11
-
Faster fHOG feature extraction / face detection
- CMake scripts now enable C++11 by default
- CMake scripts now enable C++11 by default
- Gave array2d and matrix move constructors and move assignment operators.
- Gave array2d and matrix move constructors and move assignment operators. Matrix
can also now be created from initializer lists.
</current>
</current>
<!-- ************************************************************************************** -->
<!-- ************************************************************************************** -->
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment