* add github actions from "ptheywood/cuda-cmake-github-actions"
* fix typo
* rename
* Revert "rename"
This reverts commit ded445f3bb.
* add cudnn env variable
* remove old cuda configs and try to fix installation
* add run tests
* fix test path
* actually build tests
* fix missing $
* fix syntax
* debug test path
* fix test build dir
* fix cmake syntax
* fix working directory
* Revert "debug test path"
This reverts commit 6d5b4e7a1f.
* uncomment other configurations
* fix run test path
* fix test run dir again
* syntax
* it's really not my day...
* remove incompatible configs and make tests run silently
* add initial windows script from ptheywood/cuda-cmake-github-actions
* fix windows script path
* add test for windows
* remove dot slash
* install cublas_dev
* try adding cudnn (I really need a Windows machine to test this)
* remove windows workflow scripts
* try config with ubuntu 20.10
* fix syntax error
* try to fix DLIB not using CUDA
* try again with ubuntu 20.04
* remove backticks
* remove superfluous configure step
* fix cuda parsing for v10.2
* debug test
* change working dir...
* do not run test silently
* install nvidia drivers
* only install server drivers
* load nvidia kernel module
* move module loading to script
* try to fix kernel module loading
* add simple github actions workflow
* add simple github actions workflow
* fix path?
* fix config step
* only build the dtest target
* handle Windows path
* use ./ on windows
* debugging
* debugging
* debugging
* except for the timer test on macOS, the rest works
* remove cuda custom tests, for the moment
* build in Relase mode to try to avoid timeout on Windows
* debugging
* actually change build config
* do not test the timer on macOS
* try building python
* [TYPE_SAFE_UNION] simplified some type traits and added for_each().
* added example serialization/deserialization using typeid().hash_code and different type_safe_union types
* in_place_tag is an empty struct. so don't pass const references, you're unecessarily passing 8 bytes around for now reason
* - added variant_size for type_safe_union
- added variant_alternative for type_safe_union
- removed for type_safe_union::for_each() and replaced with global function dlib::for_each_type()
* - made visit() a global
* use dlib::invoke explicitly
* - for_each_type is implemented using fold expression (or whatever the right term is) instead of template recursion. This method, in theory, yields better compile times. And if you're familiar with parameter packs, then the implementation is easier to read.
* - refactoring
- reordered function parameters in for_each_type()
- vtable implementation of apply_to_contents and visit() (sorry Davis for yet another change)
* add option to not zero out gradients and method to do it (#2477)
* Avoid different kinds of compiler warnings (#2481)
* Avoid different kinds of compiler warnings that started to appear when upgrading my build environment
* Avoid more compiler warnings
* Revert the overly verbose static_cast changes
* Make resize_bilinear and resize_bilinear_gradient take long long (previously just long)
* Circumvent what appears to be a bug in Visual Studio 2019's optimizer
(see: https://forum.juce.com/t/warning-in-the-lastest-vs2019/38267)
* Fix MSVC pragma warnings with other compilers (#2483)
* Fix warning about unused zero_gradients parameter (#2487)
* Fix warning about unused zero_gradients parameter
* match signature of other methods
* cleanup
Co-authored-by: pfeatherstone <peter@me>
Co-authored-by: Adrià Arrufat <1671644+arrufat@users.noreply.github.com>
Co-authored-by: Juha Reunanen <juha.reunanen@tomaattinen.com>
Co-authored-by: Davis King <davis@dlib.net>
* Avoid different kinds of compiler warnings that started to appear when upgrading my build environment
* Avoid more compiler warnings
* Revert the overly verbose static_cast changes
* Make resize_bilinear and resize_bilinear_gradient take long long (previously just long)
* Circumvent what appears to be a bug in Visual Studio 2019's optimizer
(see: https://forum.juce.com/t/warning-in-the-lastest-vs2019/38267)
* - added dlib::invoke_r()
- added dlib::is_invocable_r<>
- made everything constexpr and conditionally noexcept !
- added tests
This has required a refactor of dlib::invoke since std::mem_fn is not constexpr in c++11
* - !std::is_function replaced with std::is_object. Let's be a bit more precise
* - made dlib::apply constexpr and conditionally noexcept
Co-authored-by: pf <pf@me>
* Replace fc classifier with svm_multiclass_linear_trainer
* Mention about find_max_global()
Co-authored-by: Davis E. King <davis@dlib.net>
* Use double instead of float for extracted features
Co-authored-by: Davis E. King <davis@dlib.net>
* fix compilation with double features
* Revert "fix compilation with double features"
This reverts commit 76ebab4b91.
* Revert "Use double instead of float for extracted features"
This reverts commit 9a50809ebf.
* Find best C using global optimization
Co-authored-by: Davis E. King <davis@dlib.net>
* added backport of std::invoke, std::invoke_result and std::apply
* added backport of std::invoke, std::invoke_result and std::apply
* msvc doesn't like keyword 'not'
* i think this fixes detection of invoke on MSVC
* ok, i think detection of invoke stuff is fixed on windows
* - just have dlib's own implementation and don't use standard library even if c++17 is enabled.
- added tests for dlib::invoke_result_t
* added docs
* - added dlib::make_from_tuple
- added tests + docs
* - make sure you use the dlib:: namespace. Otherwise, when compiling with C++17, compiler might get confused
- use remove_reference instead of decay. That's what the standard says to use
* added dlib::is_invocable
* - defined invoke_traits. This removes dupplicate code.
- This makes absolutely no difference but is just a tiny bit nicer.
* removed the test that could potentially fail with MSVC
Co-authored-by: pfeatherstone <peter@me>
This is causing us to run out of travis-ci credits, making tests not run
at all. I deleted the duplicative tests and then disabled two additonal
ones by commenting them out that would be nice to run but I think are
not essential. In particular, the OSX one eats up a ton of credits. So
I disabled that. Maybe we can turn it back on later if we end up well
under the credit budget (or switch to github actions which appears to
have higher limits)
* wip: loss goes down when training without a dnn_trainer
if I use a dnn_trainer, it segfaults (also with bigger batch sizes...)
* remove commented code
* fix gradient computation (hopefully)
* fix loss computation
* fix crash in input_rgb_image_pair::to_tensor
* fix alias tensor offset
* refactor loss and input layers and complete the example
* add more data augmentation
* add documentation
* add documentation
* small fix in the gradient computation and reuse terms
* fix warning in comment
* use tensor_tools instead of matrix to compute the gradients
* complete the example program
* add support for mult-gpu
* Update dlib/dnn/input_abstract.h
* Update dlib/dnn/input_abstract.h
* Update dlib/dnn/loss_abstract.h
* Update examples/dnn_self_supervised_learning_ex.cpp
* Update examples/dnn_self_supervised_learning_ex.cpp
* Update examples/dnn_self_supervised_learning_ex.cpp
* Update examples/dnn_self_supervised_learning_ex.cpp
* [TYPE_SAFE_UNION] upgrade (#2443)
* [TYPE_SAFE_UNION] upgrade
* MSVC doesn't like keyword not
* MSVC doesn't like keyword and
* added tests for emplate(), copy semantics, move semantics, swap, overloaded and apply_to_contents with non void return types
* - didn't need is_void anymore
- added result_of_t
- didn't really need ostream_helper or istream_helper
- split apply_to_contents into apply_to_contents (return void) and visit (return anything so long as visitor is publicly accessible)
* - updated abstract file
* - added get_type_t
- removed deserialize_helper dupplicate
- don't use std::decay_t, that's c++14
* - removed white spaces
- don't need a return-statement when calling apply_to_contents_impl()
- use unchecked_get() whenever possible to minimise explicit use of pointer casting. lets keep that to a minimum
* - added type_safe_union_size
- added type_safe_union_size_v if C++14 is available
- added tests for above
* - test type_safe_union_size_v
* testing nested unions with visitors.
* re-added comment
* added index() in abstract file
* - refactored reset() to clear()
- added comment about clear() in abstract file
- in deserialize(), only reset the object if necessary
* - removed unecessary comment about exceptions
- removed unecessary // -------------
- struct is_valid is not mentioned in abstract. Instead rather requiring T to be a valid type, it is ensured!
- get_type and get_type_t are private. Client code shouldn't need this.
- shuffled some functions around
- type_safe_union_size and type_safe_union_size_v are removed. not needed
- reset() -> clear()
- bug fix in deserialize() index counts from 1, not 0
- improved the abstract file
* refactored index() to get_current_type_id() as per suggestion
* maybe slightly improved docs
* - HURRAY, don't need std::result_of or std::invoke_result for visit() to work. Just privately define your own type trait, in this case called return_type and return_type_t. it works!
- apply_to_contents() now always calls visit()
* example with private visitor using friendship with non-void return types.
* Fix up contracts
It can't be a post condition that T is a valid type, since the choice of T is up to the caller, it's not something these functions decide. Making it a precondition.
* Update dlib/type_safe_union/type_safe_union_kernel_abstract.h
* Update dlib/type_safe_union/type_safe_union_kernel_abstract.h
* Update dlib/type_safe_union/type_safe_union_kernel_abstract.h
* - added more tests for copy constructors/assignments, move constructors/assignments, and converting constructors/assignments
- helper_copy -> helper_forward
- added validate_type<T> in a couple of places
* - helper_move only takes non-const lvalue references. So we are not using std::move with universal references !
- use enable_if<is_valid<T>> in favor of validate_type<T>()
* - use enable_if<is_valid<T>> in favor of validate_type<T>()
* - added is_valid_check<>. This wraps enable_if<is_valid<T>,bool> and makes use of SFINAE more robust
Co-authored-by: pfeatherstone <peter@me>
Co-authored-by: pf <pf@me>
Co-authored-by: Davis E. King <davis685@gmail.com>
* Just minor cleanup of docs and renamed some stuff, tweaked formatting.
* fix spelling error
* fix most vexing parse error
Co-authored-by: Davis E. King <davis@dlib.net>
Co-authored-by: pfeatherstone <45853521+pfeatherstone@users.noreply.github.com>
Co-authored-by: pfeatherstone <peter@me>
Co-authored-by: pf <pf@me>
Co-authored-by: Davis E. King <davis685@gmail.com>
* [TYPE_SAFE_UNION] upgrade
* MSVC doesn't like keyword not
* MSVC doesn't like keyword and
* added tests for emplate(), copy semantics, move semantics, swap, overloaded and apply_to_contents with non void return types
* - didn't need is_void anymore
- added result_of_t
- didn't really need ostream_helper or istream_helper
- split apply_to_contents into apply_to_contents (return void) and visit (return anything so long as visitor is publicly accessible)
* - updated abstract file
* - added get_type_t
- removed deserialize_helper dupplicate
- don't use std::decay_t, that's c++14
* - removed white spaces
- don't need a return-statement when calling apply_to_contents_impl()
- use unchecked_get() whenever possible to minimise explicit use of pointer casting. lets keep that to a minimum
* - added type_safe_union_size
- added type_safe_union_size_v if C++14 is available
- added tests for above
* - test type_safe_union_size_v
* testing nested unions with visitors.
* re-added comment
* added index() in abstract file
* - refactored reset() to clear()
- added comment about clear() in abstract file
- in deserialize(), only reset the object if necessary
* - removed unecessary comment about exceptions
- removed unecessary // -------------
- struct is_valid is not mentioned in abstract. Instead rather requiring T to be a valid type, it is ensured!
- get_type and get_type_t are private. Client code shouldn't need this.
- shuffled some functions around
- type_safe_union_size and type_safe_union_size_v are removed. not needed
- reset() -> clear()
- bug fix in deserialize() index counts from 1, not 0
- improved the abstract file
* refactored index() to get_current_type_id() as per suggestion
* maybe slightly improved docs
* - HURRAY, don't need std::result_of or std::invoke_result for visit() to work. Just privately define your own type trait, in this case called return_type and return_type_t. it works!
- apply_to_contents() now always calls visit()
* example with private visitor using friendship with non-void return types.
* Fix up contracts
It can't be a post condition that T is a valid type, since the choice of T is up to the caller, it's not something these functions decide. Making it a precondition.
* Update dlib/type_safe_union/type_safe_union_kernel_abstract.h
* Update dlib/type_safe_union/type_safe_union_kernel_abstract.h
* Update dlib/type_safe_union/type_safe_union_kernel_abstract.h
* - added more tests for copy constructors/assignments, move constructors/assignments, and converting constructors/assignments
- helper_copy -> helper_forward
- added validate_type<T> in a couple of places
* - helper_move only takes non-const lvalue references. So we are not using std::move with universal references !
- use enable_if<is_valid<T>> in favor of validate_type<T>()
* - use enable_if<is_valid<T>> in favor of validate_type<T>()
* - added is_valid_check<>. This wraps enable_if<is_valid<T>,bool> and makes use of SFINAE more robust
Co-authored-by: pfeatherstone <peter@me>
Co-authored-by: pf <pf@me>
Co-authored-by: Davis E. King <davis685@gmail.com>
* add helper methods to implement fused convolutions
* fix grammar
* add method to disable affine layer and updated serialization
* add documentation for .disable()
* add fuse_convolutions visitor and documentation
* update docs: net is not constant
* fix xml formatting and use std::boolalpha
* fix warning and updated net requirement for visitor
* fix segfault in fuse_convolutions visitor
* copy unconditionally
* make the visitor class a friend of the con_ class
* setup the biases alias tensor after enabling bias
* simplify visitor a bit
* fix comment
* setup the biases size, somehow this got lost
* copy the parameters before resizing
* remove enable_bias() method, since the visitor is now a friend
* Revert "remove enable_bias() method, since the visitor is now a friend"
This reverts commit 35b92b1631.
* update the visitor to remove the friend requirement
* improve behavior of enable_bias
* better describe the behavior of enable_bias
* wip: use cudnncudnnConvolutionBiasActivationForward when activation has bias
* wip: fix cpu compilation
* WIP: not working fused ReLU
* WIP: forgot do disable ReLU in visitor (does not change the fact that it does not work)
* WIP: more general set of 4d tensor (still not working)
* fused convolutions seem to be working now, more testing needed
* move visitor to the bottom of the file
* fix CPU-side and code clean up
* Do not try to fuse the activation layers
Fusing the activation layers in one cuDNN call is only supported when using
the cuDNN ones (ReLU, Sigmoid, TanH...) which might lead to suprising
behavior. So, let's just fuse the batch norm and the convolution into one
cuDNN call using the IDENTITY activation function.
* Set the correct forward algorithm for the identity activation
Ref: https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnConvolutionBiasActivationForward
* move the affine alias template to its original position
* wip
* remove unused param in relu and simplify example (I will delete it before merge)
* simplify conv bias logic and fix deserialization issue
* fix enabling bias on convolutions
* remove test example
* fix typo
* update documentation
* update documentation
* remove ccache leftovers from CMakeLists.txt
* Re-add new line
* fix enable/disable bias on unallocated networks
* update comment to mention cudnnConvolutionBiasActivationForward
* fix typo
Co-authored-by: Davis E. King <davis@dlib.net>
* Apply documentation suggestions from code review
Co-authored-by: Davis E. King <davis@dlib.net>
* update affine docs to talk in terms of gamma and beta
* simplify tensor_conv interface
* fix tensor_conv operator() with biases
* add fuse_layers test
* add an example on how to use the fuse_layers function
* fix typo
Co-authored-by: Davis E. King <davis@dlib.net>
* Don't try to use labels in unsupervised losses
I hope that is the right way of fixing this...
* fix it by duplicating most code in send_job (works on my machine)
I will probably need to find a way to reuse the code
* try to fix it reusing the code... not sure though
* Revert "try to fix it reusing the code... not sure though"
This reverts commit f308cac6df.
* check the type of the training label to fix the issue instead
Davis King