Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge model compatibility fixes from 1.0rc branch. (#5305)

Browse Source 
* Port test model compatibility.
* Port logit model fix.

https://github.com/dmlc/xgboost/pull/5248
https://github.com/dmlc/xgboost/pull/5281
This commit is contained in:
Jiaming Yuan
2020-02-13 20:41:59 +08:00
committed by GitHub
parent 29eeea709a
commit 911a902835
19 changed files with 550 additions and 106 deletions
Download Patch File Download Diff File Expand all files Collapse all files

193
src/learner.cc
View File

@@ -67,19 +67,26 @@ struct LearnerModelParamLegacy : public dmlc::Parameter<LearnerModelParamLegacy>
/* \brief global bias */
bst_float base_score;
/* \brief number of features */
unsigned num_feature;
uint32_t num_feature;
/* \brief number of classes, if it is multi-class classification */
int num_class;
int32_t num_class;
/*! \brief Model contain additional properties */
int contain_extra_attrs;
int32_t contain_extra_attrs;
/*! \brief Model contain eval metrics */
int contain_eval_metrics;
int32_t contain_eval_metrics;
/*! \brief the version of XGBoost. */
uint32_t major_version;
uint32_t minor_version;
/*! \brief reserved field */
int reserved[29];
int reserved[27];
/*! \brief constructor */
LearnerModelParamLegacy() {
std::memset(this, 0, sizeof(LearnerModelParamLegacy));
base_score = 0.5f;
major_version = std::get<0>(Version::Self());
minor_version = std::get<1>(Version::Self());
static_assert(sizeof(LearnerModelParamLegacy) == 136,
"Do not change the size of this struct, as it will break binary IO.");
}
// Skip other legacy fields.
Json ToJson() const {
@@ -117,8 +124,9 @@ LearnerModelParam::LearnerModelParam(
LearnerModelParamLegacy const &user_param, float base_margin)
: base_score{base_margin}, num_feature{user_param.num_feature},
num_output_group{user_param.num_class == 0
? 1
: static_cast<uint32_t>(user_param.num_class)} {}
? 1
: static_cast<uint32_t>(user_param.num_class)}
{}
struct LearnerTrainParam : public XGBoostParameter<LearnerTrainParam> {
// data split mode, can be row, col, or none.
@@ -140,7 +148,7 @@ struct LearnerTrainParam : public XGBoostParameter<LearnerTrainParam> {
.describe("Data split mode for distributed training.");
DMLC_DECLARE_FIELD(disable_default_eval_metric)
.set_default(0)
.describe("flag to disable default metric. Set to >0 to disable");
.describe("Flag to disable default metric. Set to >0 to disable");
DMLC_DECLARE_FIELD(booster)
.set_default("gbtree")
.describe("Gradient booster used for training.");
@@ -200,6 +208,7 @@ class LearnerImpl : public Learner {
Args args = {cfg_.cbegin(), cfg_.cend()};
tparam_.UpdateAllowUnknown(args);
auto mparam_backup = mparam_;
mparam_.UpdateAllowUnknown(args);
generic_parameters_.UpdateAllowUnknown(args);
generic_parameters_.CheckDeprecated();
@@ -217,17 +226,33 @@ class LearnerImpl : public Learner {
// set seed only before the model is initialized
common::GlobalRandom().seed(generic_parameters_.seed);
// must precede configure gbm since num_features is required for gbm
this->ConfigureNumFeatures();
args = {cfg_.cbegin(), cfg_.cend()}; // renew
this->ConfigureObjective(old_tparam, &args);
this->ConfigureGBM(old_tparam, args);
this->ConfigureMetrics(args);
// Before 1.0.0, we save `base_score` into binary as a transformed value by objective.
// After 1.0.0 we save the value provided by user and keep it immutable instead. To
// keep the stability, we initialize it in binary LoadModel instead of configuration.
// Under what condition should we omit the transformation:
//
// - base_score is loaded from old binary model.
//
// What are the other possible conditions:
//
// - model loaded from new binary or JSON.
// - model is created from scratch.
// - model is configured second time due to change of parameter
if (!learner_model_param_.Initialized() || mparam_.base_score != mparam_backup.base_score) {
learner_model_param_ = LearnerModelParam(mparam_,
obj_->ProbToMargin(mparam_.base_score));
}
this->ConfigureGBM(old_tparam, args);
generic_parameters_.ConfigureGpuId(this->gbm_->UseGPU());
learner_model_param_ = LearnerModelParam(mparam_,
obj_->ProbToMargin(mparam_.base_score));
this->ConfigureMetrics(args);
this->need_configuration_ = false;
if (generic_parameters_.validate_parameters) {
@@ -337,9 +362,6 @@ class LearnerImpl : public Learner {
cache_));
gbm_->LoadModel(gradient_booster);
learner_model_param_ = LearnerModelParam(mparam_,
obj_->ProbToMargin(mparam_.base_score));
auto const& j_attributes = get<Object const>(learner.at("attributes"));
attributes_.clear();
for (auto const& kv : j_attributes) {
@@ -459,6 +481,7 @@ class LearnerImpl : public Learner {
}
if (header[0] == '{') {
// Dispatch to JSON
auto json_stream = common::FixedSizeStream(&fp);
std::string buffer;
json_stream.Take(&buffer);
@@ -471,25 +494,10 @@ class LearnerImpl : public Learner {
// read parameter
CHECK_EQ(fi->Read(&mparam_, sizeof(mparam_)), sizeof(mparam_))
<< "BoostLearner: wrong model format";
{
// backward compatibility code for compatible with old model type
// for new model, Read(&name_obj_) is suffice
uint64_t len;
CHECK_EQ(fi->Read(&len, sizeof(len)), sizeof(len));
if (len >= std::numeric_limits<unsigned>::max()) {
int gap;
CHECK_EQ(fi->Read(&gap, sizeof(gap)), sizeof(gap))
<< "BoostLearner: wrong model format";
len = len >> static_cast<uint64_t>(32UL);
}
if (len != 0) {
tparam_.objective.resize(len);
CHECK_EQ(fi->Read(&tparam_.objective[0], len), len)
<< "BoostLearner: wrong model format";
}
}
CHECK(fi->Read(&tparam_.objective)) << "BoostLearner: wrong model format";
CHECK(fi->Read(&tparam_.booster)) << "BoostLearner: wrong model format";
// duplicated code with LazyInitModel
obj_.reset(ObjFunction::Create(tparam_.objective, &generic_parameters_));
gbm_.reset(GradientBooster::Create(tparam_.booster, &generic_parameters_,
&learner_model_param_, cache_));
@@ -508,34 +516,57 @@ class LearnerImpl : public Learner {
}
attributes_ = std::map<std::string, std::string>(attr.begin(), attr.end());
}
if (tparam_.objective == "count:poisson") {
std::string max_delta_step;
fi->Read(&max_delta_step);
cfg_["max_delta_step"] = max_delta_step;
bool warn_old_model { false };
if (attributes_.find("count_poisson_max_delta_step") != attributes_.cend()) {
// Loading model from < 1.0.0, objective is not saved.
cfg_["max_delta_step"] = attributes_.at("count_poisson_max_delta_step");
attributes_.erase("count_poisson_max_delta_step");
warn_old_model = true;
} else {
warn_old_model = false;
}
if (mparam_.contain_eval_metrics != 0) {
std::vector<std::string> metr;
fi->Read(&metr);
for (auto name : metr) {
metrics_.emplace_back(Metric::Create(name, &generic_parameters_));
if (mparam_.major_version >= 1) {
learner_model_param_ = LearnerModelParam(mparam_,
obj_->ProbToMargin(mparam_.base_score));
} else {
// Before 1.0.0, base_score is saved as a transformed value, and there's no version
// attribute in the saved model.
learner_model_param_ = LearnerModelParam(mparam_, mparam_.base_score);
warn_old_model = true;
}
if (attributes_.find("objective") != attributes_.cend()) {
auto obj_str = attributes_.at("objective");
auto j_obj = Json::Load({obj_str.c_str(), obj_str.size()});
obj_->LoadConfig(j_obj);
attributes_.erase("objective");
} else {
warn_old_model = true;
}
if (attributes_.find("metrics") != attributes_.cend()) {
auto metrics_str = attributes_.at("metrics");
std::vector<std::string> names { common::Split(metrics_str, ';') };
attributes_.erase("metrics");
for (auto const& n : names) {
this->SetParam(kEvalMetric, n);
}
}
if (warn_old_model) {
LOG(WARNING) << "Loading model from XGBoost < 1.0.0, consider saving it "
"again for improved compatibility";
}
// Renew the version.
mparam_.major_version = std::get<0>(Version::Self());
mparam_.minor_version = std::get<1>(Version::Self());
cfg_["num_class"] = common::ToString(mparam_.num_class);
cfg_["num_feature"] = common::ToString(mparam_.num_feature);
auto n = tparam_.__DICT__();
cfg_.insert(n.cbegin(), n.cend());
Args args = {cfg_.cbegin(), cfg_.cend()};
generic_parameters_.UpdateAllowUnknown(args);
gbm_->Configure(args);
obj_->Configure({cfg_.begin(), cfg_.end()});
for (auto& p_metric : metrics_) {
p_metric->Configure({cfg_.begin(), cfg_.end()});
}
// copy dsplit from config since it will not run again during restore
if (tparam_.dsplit == DataSplitMode::kAuto && rabit::IsDistributed()) {
tparam_.dsplit = DataSplitMode::kRow;
@@ -552,15 +583,8 @@ class LearnerImpl : public Learner {
void SaveModel(dmlc::Stream* fo) const override {
LearnerModelParamLegacy mparam = mparam_; // make a copy to potentially modify
std::vector<std::pair<std::string, std::string> > extra_attr;
// extra attributed to be added just before saving
if (tparam_.objective == "count:poisson") {
auto it = cfg_.find("max_delta_step");
if (it != cfg_.end()) {
// write `max_delta_step` parameter as extra attribute of booster
mparam.contain_extra_attrs = 1;
extra_attr.emplace_back("count_poisson_max_delta_step", it->second);
}
}
mparam.contain_extra_attrs = 1;
{
std::vector<std::string> saved_params;
// check if rabit_bootstrap_cache were set to non zero before adding to checkpoint
@@ -577,6 +601,24 @@ class LearnerImpl : public Learner {
}
}
}
{
// Similar to JSON model IO, we save the objective.
Json j_obj { Object() };
obj_->SaveConfig(&j_obj);
std::string obj_doc;
Json::Dump(j_obj, &obj_doc);
extra_attr.emplace_back("objective", obj_doc);
}
// As of 1.0.0, JVM Package and R Package uses Save/Load model for serialization.
// Remove this part once they are ported to use actual serialization methods.
if (mparam.contain_eval_metrics != 0) {
std::stringstream os;
for (auto& ev : metrics_) {
os << ev->Name() << ";";
}
extra_attr.emplace_back("metrics", os.str());
}
fo->Write(&mparam, sizeof(LearnerModelParamLegacy));
fo->Write(tparam_.objective);
fo->Write(tparam_.booster);
@@ -587,26 +629,7 @@ class LearnerImpl : public Learner {
attr[kv.first] = kv.second;
}
fo->Write(std::vector<std::pair<std::string, std::string>>(
attr.begin(), attr.end()));
}
if (tparam_.objective == "count:poisson") {
auto it = cfg_.find("max_delta_step");
if (it != cfg_.end()) {
fo->Write(it->second);
} else {
// recover value of max_delta_step from extra attributes
auto it2 = attributes_.find("count_poisson_max_delta_step");
const std::string max_delta_step
= (it2 != attributes_.end()) ? it2->second : kMaxDeltaStepDefaultValue;
fo->Write(max_delta_step);
}
}
if (mparam.contain_eval_metrics != 0) {
std::vector<std::string> metr;
for (auto& ev : metrics_) {
metr.emplace_back(ev->Name());
}
fo->Write(metr);
attr.begin(), attr.end()));
}
}
@@ -661,11 +684,13 @@ class LearnerImpl : public Learner {
If you are loading a serialized model (like pickle in Python) generated by older
XGBoost, please export the model by calling `Booster.save_model` from that version
first, then load it back in current version. See:
first, then load it back in current version. There's a simple script for helping
the process. See:
https://xgboost.readthedocs.io/en/latest/tutorials/saving_model.html
for more details about differences between saving model and serializing.
for reference to the script, and more details about differences between saving model and
serializing.
)doc";
int64_t sz {-1};
@@ -854,7 +879,8 @@ class LearnerImpl : public Learner {
void ConfigureObjective(LearnerTrainParam const& old, Args* p_args) {
// Once binary IO is gone, NONE of these config is useful.
if (cfg_.find("num_class") != cfg_.cend() && cfg_.at("num_class") != "0") {
if (cfg_.find("num_class") != cfg_.cend() && cfg_.at("num_class") != "0" &&
tparam_.objective != "multi:softprob") {
cfg_["num_output_group"] = cfg_["num_class"];
if (atoi(cfg_["num_class"].c_str()) > 1 && cfg_.count("objective") == 0) {
tparam_.objective = "multi:softmax";
@@ -919,7 +945,6 @@ class LearnerImpl : public Learner {
}
CHECK_NE(mparam_.num_feature, 0)
<< "0 feature is supplied. Are you using raw Booster interface?";
learner_model_param_.num_feature = mparam_.num_feature;
// Remove these once binary IO is gone.
cfg_["num_feature"] = common::ToString(mparam_.num_feature);
cfg_["num_class"] = common::ToString(mparam_.num_class);