Add Accelerated Failure Time loss for survival analysis task (#4763)

* [WIP] Add lower and upper bounds on the label for survival analysis

* Update test MetaInfo.SaveLoadBinary to account for extra two fields

* Don't clear qids_ for version 2 of MetaInfo

* Add SetInfo() and GetInfo() method for lower and upper bounds

* changes to aft

* Add parameter class for AFT; use enum's to represent distribution and event type

* Add AFT metric

* changes to neg grad to grad

* changes to binomial loss

* changes to overflow

* changes to eps

* changes to code refactoring

* changes to code refactoring

* changes to code refactoring

* Re-factor survival analysis

* Remove aft namespace

* Move function bodies out of AFTNormal and AFTLogistic, to reduce clutter

* Move function bodies out of AFTLoss, to reduce clutter

* Use smart pointer to store AFTDistribution and AFTLoss

* Rename AFTNoiseDistribution enum to AFTDistributionType for clarity

The enum class was not a distribution itself but a distribution type

* Add AFTDistribution::Create() method for convenience

* changes to extreme distribution

* changes to extreme distribution

* changes to extreme

* changes to extreme distribution

* changes to left censored

* deleted cout

* changes to x,mu and sd and code refactoring

* changes to print

* changes to hessian formula in censored and uncensored

* changes to variable names and pow

* changes to Logistic Pdf

* changes to parameter

* Expose lower and upper bound labels to R package

* Use example weights; normalize log likelihood metric

* changes to CHECK

* changes to logistic hessian to standard formula

* changes to logistic formula

* Comply with coding style guideline

* Revert back Rabit submodule

* Revert dmlc-core submodule

* Comply with coding style guideline (clang-tidy)

* Fix an error in AFTLoss::Gradient()

* Add missing files to amalgamation

* Address @RAMitchell's comment: minimize future change in MetaInfo interface

* Fix lint

* Fix compilation error on 32-bit target, when size_t == bst_uint

* Allocate sufficient memory to hold extra label info

* Use OpenMP to speed up

* Fix compilation on Windows

* Address reviewer's feedback

* Add unit tests for probability distributions

* Make Metric subclass of Configurable

* Address reviewer's feedback: Configure() AFT metric

* Add a dummy test for AFT metric configuration

* Complete AFT configuration test; remove debugging print

* Rename AFT parameters

* Clarify test comment

* Add a dummy test for AFT loss for uncensored case

* Fix a bug in AFT loss for uncensored labels

* Complete unit test for AFT loss metric

* Simplify unit tests for AFT metric

* Add unit test to verify aggregate output from AFT metric

* Use EXPECT_* instead of ASSERT_*, so that we run all unit tests

* Use aft_loss_param when serializing AFTObj

This is to be consistent with AFT metric

* Add unit tests for AFT Objective

* Fix OpenMP bug; clarify semantics for shared variables used in OpenMP loops

* Add comments

* Remove AFT prefix from probability distribution; put probability distribution in separate source file

* Add comments

* Define kPI and kEulerMascheroni in probability_distribution.h

* Add probability_distribution.cc to amalgamation

* Remove unnecessary diff

* Address reviewer's feedback: define variables where they're used

* Eliminate all INFs and NANs from AFT loss and gradient

* Add demo

* Add tutorial

* Fix lint

* Use 'survival:aft' to be consistent with 'survival:cox'

* Move sample data to demo/data

* Add visual demo with 1D toy data

* Add Python tests

Co-authored-by: Philip Cho <chohyu01@cs.washington.edu>
This commit is contained in:
Avinash Barnwal
2020-03-25 16:52:51 -04:00
committed by GitHub
parent 1de36cdf1e
commit dcf439932a
21 changed files with 1789 additions and 15 deletions

View File

@@ -0,0 +1,106 @@
/*!
* Copyright 2019 by Contributors
* \file survival_metric.cc
* \brief Metrics for survival analysis
* \author Avinash Barnwal, Hyunsu Cho and Toby Hocking
*/
#include <rabit/rabit.h>
#include <xgboost/metric.h>
#include <xgboost/host_device_vector.h>
#include <dmlc/registry.h>
#include <cmath>
#include <memory>
#include <vector>
#include <limits>
#include "xgboost/json.h"
#include "../common/math.h"
#include "../common/survival_util.h"
using AFTParam = xgboost::common::AFTParam;
using AFTLoss = xgboost::common::AFTLoss;
namespace xgboost {
namespace metric {
// tag the this file, used by force static link later.
DMLC_REGISTRY_FILE_TAG(survival_metric);
/*! \brief Negative log likelihood of Accelerated Failure Time model */
struct EvalAFT : public Metric {
public:
explicit EvalAFT(const char* param) {}
void Configure(const Args& args) override {
param_.UpdateAllowUnknown(args);
loss_.reset(new AFTLoss(param_.aft_loss_distribution));
}
void SaveConfig(Json* p_out) const override {
auto& out = *p_out;
out["name"] = String(this->Name());
out["aft_loss_param"] = toJson(param_);
}
void LoadConfig(Json const& in) override {
fromJson(in["aft_loss_param"], &param_);
}
bst_float Eval(const HostDeviceVector<bst_float> &preds,
const MetaInfo &info,
bool distributed) override {
CHECK_NE(info.labels_lower_bound_.Size(), 0U)
<< "y_lower cannot be empty";
CHECK_NE(info.labels_upper_bound_.Size(), 0U)
<< "y_higher cannot be empty";
CHECK_EQ(preds.Size(), info.labels_lower_bound_.Size());
CHECK_EQ(preds.Size(), info.labels_upper_bound_.Size());
/* Compute negative log likelihood for each data point and compute weighted average */
const auto& yhat = preds.HostVector();
const auto& y_lower = info.labels_lower_bound_.HostVector();
const auto& y_upper = info.labels_upper_bound_.HostVector();
const auto& weights = info.weights_.HostVector();
const bool is_null_weight = weights.empty();
const float aft_loss_distribution_scale = param_.aft_loss_distribution_scale;
CHECK_LE(yhat.size(), static_cast<size_t>(std::numeric_limits<omp_ulong>::max()))
<< "yhat is too big";
const omp_ulong nsize = static_cast<omp_ulong>(yhat.size());
double nloglik_sum = 0.0;
double weight_sum = 0.0;
#pragma omp parallel for default(none) \
firstprivate(nsize, is_null_weight, aft_loss_distribution_scale) \
shared(weights, y_lower, y_upper, yhat) reduction(+:nloglik_sum, weight_sum)
for (omp_ulong i = 0; i < nsize; ++i) {
// If weights are empty, data is unweighted so we use 1.0 everywhere
const double w = is_null_weight ? 1.0 : weights[i];
const double loss
= loss_->Loss(y_lower[i], y_upper[i], yhat[i], aft_loss_distribution_scale);
nloglik_sum += loss;
weight_sum += w;
}
double dat[2]{nloglik_sum, weight_sum};
if (distributed) {
rabit::Allreduce<rabit::op::Sum>(dat, 2);
}
return static_cast<bst_float>(dat[0] / dat[1]);
}
const char* Name() const override {
return "aft-nloglik";
}
private:
AFTParam param_;
std::unique_ptr<AFTLoss> loss_;
};
XGBOOST_REGISTER_METRIC(AFT, "aft-nloglik")
.describe("Negative log likelihood of Accelerated Failure Time model.")
.set_body([](const char* param) { return new EvalAFT(param); });
} // namespace metric
} // namespace xgboost