[python-package] fix tmp file access problem in windows

include/LightGBM/boosting.h

@@ -136,7 +136,7 @@ public:
   * \brief Dump model to json format string
   * \return Json format string of model
   */
-  virtual std::string DumpModel() const = 0;
+  virtual std::string DumpModel(int num_iteration) const = 0;
 
   /*!
   * \brief Save model to file

include/LightGBM/c_api.h

@@ -557,12 +557,14 @@ DllExport int LGBM_BoosterSaveModel(BoosterHandle handle,
 /*!
 * \brief dump model to json
 * \param handle handle
+* \param num_iteration, <= 0 means save all
 * \param buffer_len string buffer length, if buffer_len < out_len, re-allocate buffer
 * \param out_len actual output length
 * \param out_str json format string of model, need to pre-allocate memory before call this
 * \return 0 when succeed, -1 when failure happens
 */
 DllExport int LGBM_BoosterDumpModel(BoosterHandle handle,
+  int num_iteration,
   int buffer_len,
   int64_t* out_len,
   char* out_str);

python-package/lightgbm/basic.py

@@ -9,6 +9,7 @@ import sys
 import ctypes
 import json
 from tempfile import NamedTemporaryFile
+import os
 
 import numpy as np
 import scipy.sparse
@@ -131,6 +132,22 @@ def param_dict_to_str(data):
                             % (key, type(val).__name__))
     return ' '.join(pairs)
 
+class _temp_file:
+    def __enter__(self):
+        with NamedTemporaryFile(prefix="lightgbm_tmp_", delete=True) as f:
+            self.name = f.name
+        return self
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        if os.path.isfile(self.name):
+            os.remove(self.name)
+    def readlines(self):
+        with open(self.name, "r+") as f:
+            ret = f.readlines()
+        return ret
+    def writelines(self, lines):
+        with open(self.name, "w+") as f:
+            ret = f.writelines(lines)
+
 """marco definition of data type in c_api of LightGBM"""
 C_API_DTYPE_FLOAT32 = 0
 C_API_DTYPE_FLOAT64 = 1
@@ -276,7 +293,7 @@ class _InnerPredictor(object):
         if num_iteration > self.num_total_iteration:
             num_iteration = self.num_total_iteration
         if is_str(data):
-            with NamedTemporaryFile(mode='w+') as f:
+            with _temp_file() as f:
                 _safe_call(_LIB.LGBM_BoosterPredictForFile(
                     self.handle,
                     c_str(data),
@@ -1336,7 +1353,7 @@ class Booster(object):
         return self.__deepcopy__(None)
 
     def __deepcopy__(self, _):
-        with NamedTemporaryFile(mode='w+') as f:
+        with _temp_file() as f:
             self.save_model(f.name)
             return Booster(model_file=f.name)
 
@@ -1346,7 +1363,7 @@ class Booster(object):
         this.pop('train_set', None)
         this.pop('valid_sets', None)
         if handle is not None:
-            with NamedTemporaryFile(mode='w+') as f:
+            with _temp_file() as f:
                 self.save_model(f.name)
                 this["handle"] = f.readlines()
         return this
@@ -1356,9 +1373,8 @@ class Booster(object):
         if model is not None:
             handle = ctypes.c_void_p()
             out_num_iterations = ctypes.c_int64(0)
-            with NamedTemporaryFile(mode='w+') as f:
+            with _temp_file() as f:
                 f.writelines(model)
-                f.flush()
                 _safe_call(_LIB.LGBM_BoosterCreateFromModelfile(
                     c_str(f.name),
                     ctypes.byref(out_num_iterations),
@@ -1570,27 +1586,37 @@ class Booster(object):
         filename : str
             Filename to save
         num_iteration: int
-            Number of iteration that want to save. < 0 means save all
+            Number of iteration that want to save. < 0 means save the best iteration(if have)
         """
+        if num_iteration <= 0:
+            num_iteration = self.best_iteration
         _safe_call(_LIB.LGBM_BoosterSaveModel(
             self.handle,
             num_iteration,
             c_str(filename)))
 
-    def dump_model(self):
+    def dump_model(self, num_iteration=-1):
         """
         Dump model to json format
 
+        Parameters
+        ----------
+        num_iteration: int
+            Number of iteration that want to dump. < 0 means dump to best iteration(if have)
+
         Returns
         -------
         Json format of model
         """
+        if num_iteration <= 0:
+            num_iteration = self.best_iteration
         buffer_len = 1 << 20
         tmp_out_len = ctypes.c_int64(0)
         string_buffer = ctypes.create_string_buffer(buffer_len)
         ptr_string_buffer = ctypes.c_char_p(*[ctypes.addressof(string_buffer)])
         _safe_call(_LIB.LGBM_BoosterDumpModel(
             self.handle,
+            num_iteration,
             buffer_len,
             ctypes.byref(tmp_out_len),
             ptr_string_buffer))
@@ -1601,6 +1627,7 @@ class Booster(object):
             ptr_string_buffer = ctypes.c_char_p(*[ctypes.addressof(string_buffer)])
             _safe_call(_LIB.LGBM_BoosterDumpModel(
                 self.handle,
+                num_iteration,
                 actual_len,
                 ctypes.byref(tmp_out_len),
                 ptr_string_buffer))
@@ -1616,7 +1643,7 @@ class Booster(object):
             Data source for prediction
             When data type is string, it represents the path of txt file
         num_iteration : int
-            Used iteration for prediction
+            Used iteration for prediction, < 0 means predict for best iteration(if have)
         raw_score : bool
             True for predict raw score
         pred_leaf : bool
@@ -1631,6 +1658,8 @@ class Booster(object):
         Prediction result
         """
         predictor = _InnerPredictor(booster_handle=self.handle)
+        if num_iteration <= 0:
+            num_iteration = self.best_iteration
         return predictor.predict(data, num_iteration, raw_score, pred_leaf, data_has_header, is_reshape)
 
     def _to_predictor(self):

python-package/lightgbm/engine.py

@@ -190,7 +190,7 @@ def train(params, train_set, num_boost_round=100,
     if booster.attr('best_iteration') is not None:
         booster.best_iteration = int(booster.attr('best_iteration')) + 1
     else:
-        booster.best_iteration = num_boost_round
+        booster.best_iteration = -1
     return booster
 
 

src/boosting/gbdt.cpp

@@ -245,7 +245,7 @@ bool GBDT::TrainOneIter(const score_t* gradient, const score_t* hessian, bool is
 }
 
 void GBDT::RollbackOneIter() {
-  if (iter_ == 0) { return; }
+  if (iter_ <= 0) { return; }
   int cur_iter = iter_ + num_init_iteration_ - 1;
   // reset score
   for (int curr_class = 0; curr_class < num_class_; ++curr_class) {
@@ -428,7 +428,7 @@ void GBDT::Boosting() {
     GetGradients(GetTrainingScore(&num_score), gradients_.data(), hessians_.data());
 }
 
-std::string GBDT::DumpModel() const {
+std::string GBDT::DumpModel(int num_iteration) const {
   std::stringstream str_buf;
 
   str_buf << "{";
@@ -449,7 +449,11 @@ std::string GBDT::DumpModel() const {
      << std::endl;
 
   str_buf << "\"tree_info\":[";
-  for (int i = 0; i < static_cast<int>(models_.size()); ++i) {
+  int num_used_model = static_cast<int>(models_.size());
+  if (num_iteration > 0) {
+    num_used_model = std::min(num_iteration * num_class_, num_used_model);
+  } 
+  for (int i = 0; i < num_used_model; ++i) {
     if (i > 0) {
       str_buf << ",";
     }
@@ -491,13 +495,10 @@ void GBDT::SaveModelToFile(int num_iteration, const char* filename) const {
   output_file << "feature_names=" << Common::Join(feature_names.get(), " ") << std::endl;
 
   output_file << std::endl;
-  int num_used_model = 0;
-  if (num_iteration <= 0) {
-    num_used_model = static_cast<int>(models_.size());
-  } else {
-    num_used_model = num_iteration * num_class_;
+  int num_used_model = static_cast<int>(models_.size());
+  if (num_iteration > 0) {
+    num_used_model = std::min(num_iteration * num_class_, num_used_model);
   }
-  num_used_model = std::min(num_used_model, static_cast<int>(models_.size()));
   // output tree models
   for (int i = 0; i < num_used_model; ++i) {
     output_file << "Tree=" << i << std::endl;

src/boosting/gbdt.h

@@ -148,7 +148,7 @@ public:
   * \brief Dump model to json format string
   * \return Json format string of model
   */
-  std::string DumpModel() const override;
+  std::string DumpModel(int num_iteration) const override;
 
   /*!
   * \brief Save model to file
@@ -175,7 +175,6 @@ public:
   */
   inline int LabelIdx() const override { return label_idx_; }
 
-
   /*!
   * \brief Get number of weak sub-models
   * \return Number of weak sub-models
@@ -192,13 +191,10 @@ public:
   * \brief Set number of iterations for prediction
   */
   inline void SetNumIterationForPred(int num_iteration) override {
+    num_iteration_for_pred_ = static_cast<int>(models_.size()) / num_class_;
     if (num_iteration > 0) {
-      num_iteration_for_pred_ = num_iteration;
-    } else {
-      num_iteration_for_pred_ = static_cast<int>(models_.size()) / num_class_;
+      num_iteration_for_pred_ = std::min(num_iteration, num_iteration_for_pred_);
     }
-    num_iteration_for_pred_ = std::min(num_iteration_for_pred_, 
-      static_cast<int>(models_.size()) / num_class_);
   }
 
   inline double GetLeafValue(int tree_idx, int leaf_idx) const {

src/c_api.cpp

@@ -181,8 +181,8 @@ public:
     boosting_->SaveModelToFile(num_iteration, filename);
   }
 
-  std::string DumpModel() {
-    return boosting_->DumpModel();
+  std::string DumpModel(int num_iteration) {
+    return boosting_->DumpModel(num_iteration);
   }
 
   double GetLeafValue(int tree_idx, int leaf_idx) const {
@@ -581,8 +581,7 @@ DllExport int LGBM_BoosterCreateFromModelfile(
   BoosterHandle* out) {
   API_BEGIN();
   auto ret = std::unique_ptr<Booster>(new Booster(filename));
-  *out_num_iterations = static_cast<int64_t>(ret->GetBoosting()->NumberOfTotalModel()
-    / ret->GetBoosting()->NumberOfClasses());
+  *out_num_iterations = static_cast<int64_t>(ret->GetBoosting()->GetCurrentIteration());
   *out = ret.release();
   API_END();
 }
@@ -872,12 +871,13 @@ DllExport int LGBM_BoosterSaveModel(BoosterHandle handle,
 }
 
 DllExport int LGBM_BoosterDumpModel(BoosterHandle handle,
+  int num_iteration,
   int buffer_len,
   int64_t* out_len,
   char* out_str) {
   API_BEGIN();
   Booster* ref_booster = reinterpret_cast<Booster*>(handle);
-  std::string model = ref_booster->DumpModel();
+  std::string model = ref_booster->DumpModel(num_iteration);
   *out_len = static_cast<int64_t>(model.size()) + 1;
   if (*out_len <= buffer_len) {
     std::strcpy(out_str, model.c_str());

tests/python_package_test/test_basic.py

 # coding: utf-8
 # pylint: skip-file
-import unittest, tempfile
+import unittest, tempfile, os
 import numpy as np
 from sklearn.datasets import load_breast_cancer
 from sklearn.model_selection import train_test_split
@@ -31,9 +31,11 @@ class TestBasic(unittest.TestCase):
         bst.save_model("model.txt")
         pred_from_matr = bst.predict(X_test)
         with tempfile.NamedTemporaryFile() as f:
+            tname = f.name
+        with open(tname, "w+b") as f:
             np.savetxt(f, X_test, delimiter=',')
-            f.flush()
-            pred_from_file = bst.predict(f.name)
+        pred_from_file = bst.predict(tname)
+        os.remove(tname)
         self.assertEqual(len(pred_from_matr), len(pred_from_file))
         for preds in zip(pred_from_matr, pred_from_file):
             self.assertAlmostEqual(*preds, places=5)