xgboost/include/xgboost/linear_updater.h
Jiaming Yuan f0064c07ab
Refactor configuration [Part II]. (#4577)
* Refactor configuration [Part II].

* General changes:
** Remove `Init` methods to avoid ambiguity.
** Remove `Configure(std::map<>)` to avoid redundant copying and prepare for
   parameter validation. (`std::vector` is returned from `InitAllowUnknown`).
** Add name to tree updaters for easier debugging.

* Learner changes:
** Make `LearnerImpl` the only source of configuration.

    All configurations are stored and carried out by `LearnerImpl::Configure()`.

** Remove booster in C API.

    Originally kept for "compatibility reason", but did not state why.  So here
    we just remove it.

** Add a `metric_names_` field in `LearnerImpl`.
** Remove `LazyInit`.  Configuration will always be lazy.
** Run `Configure` before every iteration.

* Predictor changes:
** Allocate both cpu and gpu predictor.
** Remove cpu_predictor from gpu_predictor.

    `GBTree` is now used to dispatch the predictor.

** Remove some GPU Predictor tests.

* IO

No IO changes.  The binary model format stability is tested by comparing
hashing value of save models between two commits
2019-07-20 08:34:56 -04:00

71 lines
2.1 KiB
C++

/*
* Copyright 2018 by Contributors
*/
#pragma once
#include <dmlc/registry.h>
#include <xgboost/base.h>
#include <xgboost/data.h>
#include <xgboost/generic_parameters.h>
#include <functional>
#include <string>
#include <utility>
#include <vector>
#include "../../src/gbm/gblinear_model.h"
#include "../../src/common/host_device_vector.h"
namespace xgboost {
/*!
* \brief interface of linear updater
*/
class LinearUpdater {
protected:
GenericParameter const* learner_param_;
public:
/*! \brief virtual destructor */
virtual ~LinearUpdater() = default;
/*!
* \brief Initialize the updater with given arguments.
* \param args arguments to the objective function.
*/
virtual void Configure(
const std::vector<std::pair<std::string, std::string> >& args) = 0;
/**
* \brief Updates linear model given gradients.
*
* \param in_gpair The gradient pair statistics of the data.
* \param data Input data matrix.
* \param model Model to be updated.
* \param sum_instance_weight The sum instance weights, used to normalise l1/l2 penalty.
*/
virtual void Update(HostDeviceVector<GradientPair>* in_gpair, DMatrix* data,
gbm::GBLinearModel* model,
double sum_instance_weight) = 0;
/*!
* \brief Create a linear updater given name
* \param name Name of the linear updater.
*/
static LinearUpdater* Create(const std::string& name, GenericParameter const*);
};
/*!
* \brief Registry entry for linear updater.
*/
struct LinearUpdaterReg
: public dmlc::FunctionRegEntryBase<LinearUpdaterReg,
std::function<LinearUpdater*()> > {};
/*!
* \brief Macro to register linear updater.
*/
#define XGBOOST_REGISTER_LINEAR_UPDATER(UniqueId, Name) \
static DMLC_ATTRIBUTE_UNUSED ::xgboost::LinearUpdaterReg& \
__make_##LinearUpdaterReg##_##UniqueId##__ = \
::dmlc::Registry< ::xgboost::LinearUpdaterReg>::Get()->__REGISTER__( \
Name)
} // namespace xgboost