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

Clang-tidy static analysis (#3222)

Browse Source 
* Clang-tidy static analysis

* Modernise checks

* Google coding standard checks

* Identifier renaming according to Google style
This commit is contained in:
Rory Mitchell
2018-04-19 18:57:13 +12:00
committed by GitHub
parent 3242b0a378
commit ccf80703ef
97 changed files with 3407 additions and 3354 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
src/metric/elementwise_metric.cc
View File

@@ -24,16 +24,16 @@ struct EvalEWiseBase : public Metric {
bst_float Eval(const std::vector<bst_float>& preds,
const MetaInfo& info,
bool distributed) const override {
CHECK_NE(info.labels.size(), 0U) << "label set cannot be empty";
CHECK_EQ(preds.size(), info.labels.size())
CHECK_NE(info.labels_.size(), 0U) << "label set cannot be empty";
CHECK_EQ(preds.size(), info.labels_.size())
<< "label and prediction size not match, "
<< "hint: use merror or mlogloss for multi-class classification";
const omp_ulong ndata = static_cast<omp_ulong>(info.labels.size());
const auto ndata = static_cast<omp_ulong>(info.labels_.size());
double sum = 0.0, wsum = 0.0;
#pragma omp parallel for reduction(+: sum, wsum) schedule(static)
for (omp_ulong i = 0; i < ndata; ++i) {
const bst_float wt = info.GetWeight(i);
sum += static_cast<const Derived*>(this)->EvalRow(info.labels[i], preds[i]) * wt;
sum += static_cast<const Derived*>(this)->EvalRow(info.labels_[i], preds[i]) * wt;
wsum += wt;
}
double dat[2]; dat[0] = sum, dat[1] = wsum;

14
src/metric/multiclass_metric.cc
View File

@@ -23,23 +23,23 @@ struct EvalMClassBase : public Metric {
bst_float Eval(const std::vector<bst_float> &preds,
const MetaInfo &info,
bool distributed) const override {
CHECK_NE(info.labels.size(), 0U) << "label set cannot be empty";
CHECK(preds.size() % info.labels.size() == 0)
CHECK_NE(info.labels_.size(), 0U) << "label set cannot be empty";
CHECK(preds.size() % info.labels_.size() == 0)
<< "label and prediction size not match";
const size_t nclass = preds.size() / info.labels.size();
const size_t nclass = preds.size() / info.labels_.size();
CHECK_GE(nclass, 1U)
<< "mlogloss and merror are only used for multi-class classification,"
<< " use logloss for binary classification";
const bst_omp_uint ndata = static_cast<bst_omp_uint>(info.labels.size());
const auto ndata = static_cast<bst_omp_uint>(info.labels_.size());
double sum = 0.0, wsum = 0.0;
int label_error = 0;
#pragma omp parallel for reduction(+: sum, wsum) schedule(static)
for (bst_omp_uint i = 0; i < ndata; ++i) {
const bst_float wt = info.GetWeight(i);
int label = static_cast<int>(info.labels[i]);
auto label = static_cast<int>(info.labels_[i]);
if (label >= 0 && label < static_cast<int>(nclass)) {
sum += Derived::EvalRow(label,
dmlc::BeginPtr(preds) + i * nclass,
preds.data() + i * nclass,
nclass) * wt;
wsum += wt;
} else {
@@ -99,7 +99,7 @@ struct EvalMultiLogLoss : public EvalMClassBase<EvalMultiLogLoss> {
const bst_float *pred,
size_t nclass) {
const bst_float eps = 1e-16f;
size_t k = static_cast<size_t>(label);
auto k = static_cast<size_t>(label);
if (pred[k] > eps) {
return -std::log(pred[k]);
} else {

67
src/metric/rank_metric.cc
View File

@@ -19,7 +19,7 @@ DMLC_REGISTRY_FILE_TAG(rank_metric);
struct EvalAMS : public Metric {
public:
explicit EvalAMS(const char* param) {
CHECK(param != nullptr)
CHECK(param != nullptr) // NOLINT
<< "AMS must be in format ams@k";
ratio_ = atof(param);
std::ostringstream os;
@@ -32,7 +32,7 @@ struct EvalAMS : public Metric {
CHECK(!distributed) << "metric AMS do not support distributed evaluation";
using namespace std; // NOLINT(*)
const bst_omp_uint ndata = static_cast<bst_omp_uint>(info.labels.size());
const auto ndata = static_cast<bst_omp_uint>(info.labels_.size());
std::vector<std::pair<bst_float, unsigned> > rec(ndata);
#pragma omp parallel for schedule(static)
@@ -40,7 +40,7 @@ struct EvalAMS : public Metric {
rec[i] = std::make_pair(preds[i], i);
}
std::sort(rec.begin(), rec.end(), common::CmpFirst);
unsigned ntop = static_cast<unsigned>(ratio_ * ndata);
auto ntop = static_cast<unsigned>(ratio_ * ndata);
if (ntop == 0) ntop = ndata;
const double br = 10.0;
unsigned thresindex = 0;
@@ -48,7 +48,7 @@ struct EvalAMS : public Metric {
for (unsigned i = 0; i < static_cast<unsigned>(ndata-1) && i < ntop; ++i) {
const unsigned ridx = rec[i].second;
const bst_float wt = info.GetWeight(ridx);
if (info.labels[ridx] > 0.5f) {
if (info.labels_[ridx] > 0.5f) {
s_tp += wt;
} else {
b_fp += wt;
@@ -84,16 +84,16 @@ struct EvalAuc : public Metric {
bst_float Eval(const std::vector<bst_float> &preds,
const MetaInfo &info,
bool distributed) const override {
CHECK_NE(info.labels.size(), 0U) << "label set cannot be empty";
CHECK_EQ(preds.size(), info.labels.size())
CHECK_NE(info.labels_.size(), 0U) << "label set cannot be empty";
CHECK_EQ(preds.size(), info.labels_.size())
<< "label size predict size not match";
std::vector<unsigned> tgptr(2, 0);
tgptr[1] = static_cast<unsigned>(info.labels.size());
tgptr[1] = static_cast<unsigned>(info.labels_.size());
const std::vector<unsigned> &gptr = info.group_ptr.size() == 0 ? tgptr : info.group_ptr;
CHECK_EQ(gptr.back(), info.labels.size())
const std::vector<unsigned> &gptr = info.group_ptr_.size() == 0 ? tgptr : info.group_ptr_;
CHECK_EQ(gptr.back(), info.labels_.size())
<< "EvalAuc: group structure must match number of prediction";
const bst_omp_uint ngroup = static_cast<bst_omp_uint>(gptr.size() - 1);
const auto ngroup = static_cast<bst_omp_uint>(gptr.size() - 1);
// sum statistics
bst_float sum_auc = 0.0f;
int auc_error = 0;
@@ -102,7 +102,7 @@ struct EvalAuc : public Metric {
for (bst_omp_uint k = 0; k < ngroup; ++k) {
rec.clear();
for (unsigned j = gptr[k]; j < gptr[k + 1]; ++j) {
rec.push_back(std::make_pair(preds[j], j));
rec.emplace_back(preds[j], j);
}
XGBOOST_PARALLEL_SORT(rec.begin(), rec.end(), common::CmpFirst);
// calculate AUC
@@ -110,7 +110,7 @@ struct EvalAuc : public Metric {
double sum_npos = 0.0, sum_nneg = 0.0, buf_pos = 0.0, buf_neg = 0.0;
for (size_t j = 0; j < rec.size(); ++j) {
const bst_float wt = info.GetWeight(rec[j].second);
const bst_float ctr = info.labels[rec[j].second];
const bst_float ctr = info.labels_[rec[j].second];
// keep bucketing predictions in same bucket
if (j != 0 && rec[j].first != rec[j - 1].first) {
sum_pospair += buf_neg * (sum_npos + buf_pos *0.5);
@@ -156,16 +156,16 @@ struct EvalRankList : public Metric {
bst_float Eval(const std::vector<bst_float> &preds,
const MetaInfo &info,
bool distributed) const override {
CHECK_EQ(preds.size(), info.labels.size())
CHECK_EQ(preds.size(), info.labels_.size())
<< "label size predict size not match";
// quick consistency when group is not available
std::vector<unsigned> tgptr(2, 0);
tgptr[1] = static_cast<unsigned>(preds.size());
const std::vector<unsigned> &gptr = info.group_ptr.size() == 0 ? tgptr : info.group_ptr;
const std::vector<unsigned> &gptr = info.group_ptr_.size() == 0 ? tgptr : info.group_ptr_;
CHECK_NE(gptr.size(), 0U) << "must specify group when constructing rank file";
CHECK_EQ(gptr.back(), preds.size())
<< "EvalRanklist: group structure must match number of prediction";
const bst_omp_uint ngroup = static_cast<bst_omp_uint>(gptr.size() - 1);
const auto ngroup = static_cast<bst_omp_uint>(gptr.size() - 1);
// sum statistics
double sum_metric = 0.0f;
#pragma omp parallel reduction(+:sum_metric)
@@ -176,7 +176,7 @@ struct EvalRankList : public Metric {
for (bst_omp_uint k = 0; k < ngroup; ++k) {
rec.clear();
for (unsigned j = gptr[k]; j < gptr[k + 1]; ++j) {
rec.push_back(std::make_pair(preds[j], static_cast<int>(info.labels[j])));
rec.emplace_back(preds[j], static_cast<int>(info.labels_[j]));
}
sum_metric += this->EvalMetric(rec);
}
@@ -230,7 +230,7 @@ struct EvalPrecision : public EvalRankList{
explicit EvalPrecision(const char *name) : EvalRankList("pre", name) {}
protected:
virtual bst_float EvalMetric(std::vector< std::pair<bst_float, unsigned> > &rec) const {
bst_float EvalMetric(std::vector< std::pair<bst_float, unsigned> > &rec) const override {
// calculate Precision
std::sort(rec.begin(), rec.end(), common::CmpFirst);
unsigned nhit = 0;
@@ -279,7 +279,7 @@ struct EvalMAP : public EvalRankList {
explicit EvalMAP(const char *name) : EvalRankList("map", name) {}
protected:
virtual bst_float EvalMetric(std::vector< std::pair<bst_float, unsigned> > &rec) const {
bst_float EvalMetric(std::vector< std::pair<bst_float, unsigned> > &rec) const override {
std::sort(rec.begin(), rec.end(), common::CmpFirst);
unsigned nhits = 0;
double sumap = 0.0;
@@ -307,14 +307,14 @@ struct EvalMAP : public EvalRankList {
/*! \brief Cox: Partial likelihood of the Cox proportional hazards model */
struct EvalCox : public Metric {
public:
EvalCox() {}
EvalCox() = default;
bst_float Eval(const std::vector<bst_float> &preds,
const MetaInfo &info,
bool distributed) const override {
CHECK(!distributed) << "Cox metric does not support distributed evaluation";
using namespace std; // NOLINT(*)
const bst_omp_uint ndata = static_cast<bst_omp_uint>(info.labels.size());
const auto ndata = static_cast<bst_omp_uint>(info.labels_.size());
const std::vector<size_t> &label_order = info.LabelAbsSort();
// pre-compute a sum for the denominator
@@ -328,7 +328,7 @@ struct EvalCox : public Metric {
bst_omp_uint num_events = 0;
for (bst_omp_uint i = 0; i < ndata; ++i) {
const size_t ind = label_order[i];
const auto label = info.labels[ind];
const auto label = info.labels_[ind];
if (label > 0) {
out -= log(preds[ind]) - log(exp_p_sum);
++num_events;
@@ -336,7 +336,7 @@ struct EvalCox : public Metric {
// only update the denominator after we move forward in time (labels are sorted)
accumulated_sum += preds[ind];
if (i == ndata - 1 || std::abs(label) < std::abs(info.labels[label_order[i + 1]])) {
if (i == ndata - 1 || std::abs(label) < std::abs(info.labels_[label_order[i + 1]])) {
exp_p_sum -= accumulated_sum;
accumulated_sum = 0;
}
@@ -358,16 +358,16 @@ struct EvalAucPR : public Metric {
bst_float Eval(const std::vector<bst_float> &preds, const MetaInfo &info,
bool distributed) const override {
CHECK_NE(info.labels.size(), 0U) << "label set cannot be empty";
CHECK_EQ(preds.size(), info.labels.size())
CHECK_NE(info.labels_.size(), 0U) << "label set cannot be empty";
CHECK_EQ(preds.size(), info.labels_.size())
<< "label size predict size not match";
std::vector<unsigned> tgptr(2, 0);
tgptr[1] = static_cast<unsigned>(info.labels.size());
tgptr[1] = static_cast<unsigned>(info.labels_.size());
const std::vector<unsigned> &gptr =
info.group_ptr.size() == 0 ? tgptr : info.group_ptr;
CHECK_EQ(gptr.back(), info.labels.size())
info.group_ptr_.size() == 0 ? tgptr : info.group_ptr_;
CHECK_EQ(gptr.back(), info.labels_.size())
<< "EvalAucPR: group structure must match number of prediction";
const bst_omp_uint ngroup = static_cast<bst_omp_uint>(gptr.size() - 1);
const auto ngroup = static_cast<bst_omp_uint>(gptr.size() - 1);
// sum statistics
double auc = 0.0;
int auc_error = 0, auc_gt_one = 0;
@@ -378,9 +378,9 @@ struct EvalAucPR : public Metric {
double total_neg = 0.0;
rec.clear();
for (unsigned j = gptr[k]; j < gptr[k + 1]; ++j) {
total_pos += info.GetWeight(j) * info.labels[j];
total_neg += info.GetWeight(j) * (1.0f - info.labels[j]);
rec.push_back(std::make_pair(preds[j], j));
total_pos += info.GetWeight(j) * info.labels_[j];
total_neg += info.GetWeight(j) * (1.0f - info.labels_[j]);
rec.emplace_back(preds[j], j);
}
XGBOOST_PARALLEL_SORT(rec.begin(), rec.end(), common::CmpFirst);
// we need pos > 0 && neg > 0
@@ -390,11 +390,10 @@ struct EvalAucPR : public Metric {
// calculate AUC
double tp = 0.0, prevtp = 0.0, fp = 0.0, prevfp = 0.0, h = 0.0, a = 0.0, b = 0.0;
for (size_t j = 0; j < rec.size(); ++j) {
tp += info.GetWeight(rec[j].second) * info.labels[rec[j].second];
fp += info.GetWeight(rec[j].second) * (1.0f - info.labels[rec[j].second]);
tp += info.GetWeight(rec[j].second) * info.labels_[rec[j].second];
fp += info.GetWeight(rec[j].second) * (1.0f - info.labels_[rec[j].second]);
if ((j < rec.size() - 1 && rec[j].first != rec[j + 1].first) || j == rec.size() - 1) {
if (tp == prevtp) {
h = 1.0;
a = 1.0;
b = 0.0;
} else {