basic.py 133 KB
Newer Older
wxchan's avatar
wxchan committed
1
# coding: utf-8
2
"""Wrapper for C API of LightGBM."""
3
from __future__ import absolute_import, print_function
wxchan's avatar
wxchan committed
4

5
import copy
wxchan's avatar
wxchan committed
6
import ctypes
7
import os
wxchan's avatar
wxchan committed
8
import warnings
wxchan's avatar
wxchan committed
9
from tempfile import NamedTemporaryFile
10
from collections import OrderedDict
wxchan's avatar
wxchan committed
11
12
13
14

import numpy as np
import scipy.sparse

15
from .compat import (PANDAS_INSTALLED, DataFrame, Series, is_dtype_sparse,
16
                     DataTable,
17
18
                     decode_string, string_type,
                     integer_types, numeric_types,
19
                     json, json_default_with_numpy,
20
                     range_, zip_)
wxchan's avatar
wxchan committed
21
22
from .libpath import find_lib_path

wxchan's avatar
wxchan committed
23

24
25
26
27
28
def _log_callback(msg):
    """Redirect logs from native library into Python console."""
    print("{0:s}".format(decode_string(msg)), end='')


wxchan's avatar
wxchan committed
29
def _load_lib():
30
    """Load LightGBM library."""
wxchan's avatar
wxchan committed
31
32
    lib_path = find_lib_path()
    if len(lib_path) == 0:
33
        return None
wxchan's avatar
wxchan committed
34
35
    lib = ctypes.cdll.LoadLibrary(lib_path[0])
    lib.LGBM_GetLastError.restype = ctypes.c_char_p
36
37
38
39
    callback = ctypes.CFUNCTYPE(None, ctypes.c_char_p)
    lib.callback = callback(_log_callback)
    if lib.LGBM_RegisterLogCallback(lib.callback) != 0:
        raise LightGBMError(decode_string(lib.LGBM_GetLastError()))
wxchan's avatar
wxchan committed
40
41
    return lib

wxchan's avatar
wxchan committed
42

wxchan's avatar
wxchan committed
43
44
_LIB = _load_lib()

wxchan's avatar
wxchan committed
45

wxchan's avatar
wxchan committed
46
def _safe_call(ret):
47
48
    """Check the return value from C API call.

wxchan's avatar
wxchan committed
49
50
51
    Parameters
    ----------
    ret : int
52
        The return value from C API calls.
wxchan's avatar
wxchan committed
53
54
    """
    if ret != 0:
55
        raise LightGBMError(decode_string(_LIB.LGBM_GetLastError()))
wxchan's avatar
wxchan committed
56

wxchan's avatar
wxchan committed
57

wxchan's avatar
wxchan committed
58
def is_numeric(obj):
59
    """Check whether object is a number or not, include numpy number, etc."""
wxchan's avatar
wxchan committed
60
61
62
    try:
        float(obj)
        return True
wxchan's avatar
wxchan committed
63
64
65
    except (TypeError, ValueError):
        # TypeError: obj is not a string or a number
        # ValueError: invalid literal
wxchan's avatar
wxchan committed
66
67
        return False

wxchan's avatar
wxchan committed
68

wxchan's avatar
wxchan committed
69
def is_numpy_1d_array(data):
70
    """Check whether data is a numpy 1-D array."""
71
    return isinstance(data, np.ndarray) and len(data.shape) == 1
wxchan's avatar
wxchan committed
72

wxchan's avatar
wxchan committed
73

wxchan's avatar
wxchan committed
74
def is_1d_list(data):
75
76
    """Check whether data is a 1-D list."""
    return isinstance(data, list) and (not data or is_numeric(data[0]))
wxchan's avatar
wxchan committed
77

wxchan's avatar
wxchan committed
78

79
def list_to_1d_numpy(data, dtype=np.float32, name='list'):
80
    """Convert data to numpy 1-D array."""
wxchan's avatar
wxchan committed
81
82
83
84
85
86
87
    if is_numpy_1d_array(data):
        if data.dtype == dtype:
            return data
        else:
            return data.astype(dtype=dtype, copy=False)
    elif is_1d_list(data):
        return np.array(data, dtype=dtype, copy=False)
88
    elif isinstance(data, Series):
89
90
        if _get_bad_pandas_dtypes([data.dtypes]):
            raise ValueError('Series.dtypes must be int, float or bool')
91
        return np.array(data, dtype=dtype, copy=False)  # SparseArray should be supported as well
wxchan's avatar
wxchan committed
92
    else:
93
94
        raise TypeError("Wrong type({0}) for {1}.\n"
                        "It should be list, numpy 1-D array or pandas Series".format(type(data).__name__, name))
wxchan's avatar
wxchan committed
95

wxchan's avatar
wxchan committed
96

wxchan's avatar
wxchan committed
97
def cfloat32_array_to_numpy(cptr, length):
98
    """Convert a ctypes float pointer array to a numpy array."""
wxchan's avatar
wxchan committed
99
    if isinstance(cptr, ctypes.POINTER(ctypes.c_float)):
100
        return np.fromiter(cptr, dtype=np.float32, count=length)
wxchan's avatar
wxchan committed
101
    else:
102
        raise RuntimeError('Expected float pointer')
wxchan's avatar
wxchan committed
103

Guolin Ke's avatar
Guolin Ke committed
104

Guolin Ke's avatar
Guolin Ke committed
105
def cfloat64_array_to_numpy(cptr, length):
106
    """Convert a ctypes double pointer array to a numpy array."""
Guolin Ke's avatar
Guolin Ke committed
107
108
109
110
111
    if isinstance(cptr, ctypes.POINTER(ctypes.c_double)):
        return np.fromiter(cptr, dtype=np.float64, count=length)
    else:
        raise RuntimeError('Expected double pointer')

wxchan's avatar
wxchan committed
112

wxchan's avatar
wxchan committed
113
def cint32_array_to_numpy(cptr, length):
114
    """Convert a ctypes int pointer array to a numpy array."""
wxchan's avatar
wxchan committed
115
    if isinstance(cptr, ctypes.POINTER(ctypes.c_int32)):
116
        return np.fromiter(cptr, dtype=np.int32, count=length)
wxchan's avatar
wxchan committed
117
    else:
118
119
120
121
122
123
124
125
126
        raise RuntimeError('Expected int32 pointer')


def cint64_array_to_numpy(cptr, length):
    """Convert a ctypes int pointer array to a numpy array."""
    if isinstance(cptr, ctypes.POINTER(ctypes.c_int64)):
        return np.fromiter(cptr, dtype=np.int64, count=length)
    else:
        raise RuntimeError('Expected int64 pointer')
wxchan's avatar
wxchan committed
127

wxchan's avatar
wxchan committed
128

wxchan's avatar
wxchan committed
129
def c_str(string):
130
    """Convert a Python string to C string."""
wxchan's avatar
wxchan committed
131
132
    return ctypes.c_char_p(string.encode('utf-8'))

wxchan's avatar
wxchan committed
133

wxchan's avatar
wxchan committed
134
def c_array(ctype, values):
135
    """Convert a Python array to C array."""
wxchan's avatar
wxchan committed
136
137
    return (ctype * len(values))(*values)

wxchan's avatar
wxchan committed
138

wxchan's avatar
wxchan committed
139
def param_dict_to_str(data):
140
    """Convert Python dictionary to string, which is passed to C API."""
141
    if data is None or not data:
wxchan's avatar
wxchan committed
142
143
144
        return ""
    pairs = []
    for key, val in data.items():
145
        if isinstance(val, (list, tuple, set)) or is_numpy_1d_array(val):
146
147
148
149
150
151
            def to_string(x):
                if isinstance(x, list):
                    return "[{}]".format(','.join(map(str, x)))
                else:
                    return str(x)
            pairs.append(str(key) + '=' + ','.join(map(to_string, val)))
wxchan's avatar
wxchan committed
152
        elif isinstance(val, string_type) or isinstance(val, numeric_types) or is_numeric(val):
wxchan's avatar
wxchan committed
153
            pairs.append(str(key) + '=' + str(val))
154
        elif val is not None:
155
            raise TypeError('Unknown type of parameter:%s, got:%s'
wxchan's avatar
wxchan committed
156
157
                            % (key, type(val).__name__))
    return ' '.join(pairs)
158

wxchan's avatar
wxchan committed
159

160
class _TempFile(object):
161
162
163
164
    def __enter__(self):
        with NamedTemporaryFile(prefix="lightgbm_tmp_", delete=True) as f:
            self.name = f.name
        return self
wxchan's avatar
wxchan committed
165

166
167
168
    def __exit__(self, exc_type, exc_val, exc_tb):
        if os.path.isfile(self.name):
            os.remove(self.name)
wxchan's avatar
wxchan committed
169

170
171
172
173
    def readlines(self):
        with open(self.name, "r+") as f:
            ret = f.readlines()
        return ret
wxchan's avatar
wxchan committed
174

175
176
    def writelines(self, lines):
        with open(self.name, "w+") as f:
177
            f.writelines(lines)
178

wxchan's avatar
wxchan committed
179

180
class LightGBMError(Exception):
181
182
    """Error thrown by LightGBM."""

183
184
185
    pass


186
class _ConfigAliases(object):
187
188
189
    aliases = {"bin_construct_sample_cnt": {"bin_construct_sample_cnt",
                                            "subsample_for_bin"},
               "boosting": {"boosting",
190
191
192
193
194
195
                            "boosting_type",
                            "boost"},
               "categorical_feature": {"categorical_feature",
                                       "cat_feature",
                                       "categorical_column",
                                       "cat_column"},
196
197
               "data_random_seed": {"data_random_seed",
                                    "data_seed"},
198
199
200
201
               "early_stopping_round": {"early_stopping_round",
                                        "early_stopping_rounds",
                                        "early_stopping",
                                        "n_iter_no_change"},
202
203
204
               "enable_bundle": {"enable_bundle",
                                 "is_enable_bundle",
                                 "bundle"},
205
206
207
208
209
               "eval_at": {"eval_at",
                           "ndcg_eval_at",
                           "ndcg_at",
                           "map_eval_at",
                           "map_at"},
210
211
212
213
214
215
               "group_column": {"group_column",
                                "group",
                                "group_id",
                                "query_column",
                                "query",
                                "query_id"},
216
217
               "header": {"header",
                          "has_header"},
218
219
220
221
222
223
224
225
226
               "ignore_column": {"ignore_column",
                                 "ignore_feature",
                                 "blacklist"},
               "is_enable_sparse": {"is_enable_sparse",
                                    "is_sparse",
                                    "enable_sparse",
                                    "sparse"},
               "label_column": {"label_column",
                                "label"},
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
               "machines": {"machines",
                            "workers",
                            "nodes"},
               "metric": {"metric",
                          "metrics",
                          "metric_types"},
               "num_class": {"num_class",
                             "num_classes"},
               "num_iterations": {"num_iterations",
                                  "num_iteration",
                                  "n_iter",
                                  "num_tree",
                                  "num_trees",
                                  "num_round",
                                  "num_rounds",
                                  "num_boost_round",
                                  "n_estimators"},
               "objective": {"objective",
                             "objective_type",
                             "app",
                             "application"},
248
249
250
251
252
               "pre_partition": {"pre_partition",
                                 "is_pre_partition"},
               "two_round": {"two_round",
                             "two_round_loading",
                             "use_two_round_loading"},
253
               "verbosity": {"verbosity",
254
255
256
                             "verbose"},
               "weight_column": {"weight_column",
                                 "weight"}}
257
258
259
260
261

    @classmethod
    def get(cls, *args):
        ret = set()
        for i in args:
262
            ret |= cls.aliases.get(i, {i})
263
264
265
        return ret


266
267
MAX_INT32 = (1 << 31) - 1

268
"""Macro definition of data type in C API of LightGBM"""
wxchan's avatar
wxchan committed
269
270
271
272
C_API_DTYPE_FLOAT32 = 0
C_API_DTYPE_FLOAT64 = 1
C_API_DTYPE_INT32 = 2
C_API_DTYPE_INT64 = 3
Guolin Ke's avatar
Guolin Ke committed
273

274
"""Matrix is row major in Python"""
wxchan's avatar
wxchan committed
275
276
C_API_IS_ROW_MAJOR = 1

277
"""Macro definition of prediction type in C API of LightGBM"""
wxchan's avatar
wxchan committed
278
279
280
C_API_PREDICT_NORMAL = 0
C_API_PREDICT_RAW_SCORE = 1
C_API_PREDICT_LEAF_INDEX = 2
281
C_API_PREDICT_CONTRIB = 3
wxchan's avatar
wxchan committed
282

283
284
285
286
"""Macro definition of sparse matrix type"""
C_API_MATRIX_TYPE_CSR = 0
C_API_MATRIX_TYPE_CSC = 1

287
288
289
290
"""Macro definition of feature importance type"""
C_API_FEATURE_IMPORTANCE_SPLIT = 0
C_API_FEATURE_IMPORTANCE_GAIN = 1

291
"""Data type of data field"""
wxchan's avatar
wxchan committed
292
293
FIELD_TYPE_MAPPER = {"label": C_API_DTYPE_FLOAT32,
                     "weight": C_API_DTYPE_FLOAT32,
Guolin Ke's avatar
Guolin Ke committed
294
                     "init_score": C_API_DTYPE_FLOAT64,
295
                     "group": C_API_DTYPE_INT32}
wxchan's avatar
wxchan committed
296

297
298
299
300
"""String name to int feature importance type mapper"""
FEATURE_IMPORTANCE_TYPE_MAPPER = {"split": C_API_FEATURE_IMPORTANCE_SPLIT,
                                  "gain": C_API_FEATURE_IMPORTANCE_GAIN}

wxchan's avatar
wxchan committed
301

302
def convert_from_sliced_object(data):
303
    """Fix the memory of multi-dimensional sliced object."""
304
    if isinstance(data, np.ndarray) and isinstance(data.base, np.ndarray):
305
        if not data.flags.c_contiguous:
306
307
            warnings.warn("Usage of np.ndarray subset (sliced data) is not recommended "
                          "due to it will double the peak memory cost in LightGBM.")
308
309
310
311
            return np.copy(data)
    return data


wxchan's avatar
wxchan committed
312
def c_float_array(data):
313
    """Get pointer of float numpy array / list."""
wxchan's avatar
wxchan committed
314
315
316
    if is_1d_list(data):
        data = np.array(data, copy=False)
    if is_numpy_1d_array(data):
317
318
        data = convert_from_sliced_object(data)
        assert data.flags.c_contiguous
wxchan's avatar
wxchan committed
319
320
321
322
323
324
325
        if data.dtype == np.float32:
            ptr_data = data.ctypes.data_as(ctypes.POINTER(ctypes.c_float))
            type_data = C_API_DTYPE_FLOAT32
        elif data.dtype == np.float64:
            ptr_data = data.ctypes.data_as(ctypes.POINTER(ctypes.c_double))
            type_data = C_API_DTYPE_FLOAT64
        else:
326
            raise TypeError("Expected np.float32 or np.float64, met type({})"
wxchan's avatar
wxchan committed
327
328
                            .format(data.dtype))
    else:
329
        raise TypeError("Unknown type({})".format(type(data).__name__))
330
    return (ptr_data, type_data, data)  # return `data` to avoid the temporary copy is freed
wxchan's avatar
wxchan committed
331

wxchan's avatar
wxchan committed
332

wxchan's avatar
wxchan committed
333
def c_int_array(data):
334
    """Get pointer of int numpy array / list."""
wxchan's avatar
wxchan committed
335
336
337
    if is_1d_list(data):
        data = np.array(data, copy=False)
    if is_numpy_1d_array(data):
338
339
        data = convert_from_sliced_object(data)
        assert data.flags.c_contiguous
wxchan's avatar
wxchan committed
340
341
342
343
344
345
346
        if data.dtype == np.int32:
            ptr_data = data.ctypes.data_as(ctypes.POINTER(ctypes.c_int32))
            type_data = C_API_DTYPE_INT32
        elif data.dtype == np.int64:
            ptr_data = data.ctypes.data_as(ctypes.POINTER(ctypes.c_int64))
            type_data = C_API_DTYPE_INT64
        else:
347
            raise TypeError("Expected np.int32 or np.int64, met type({})"
wxchan's avatar
wxchan committed
348
349
                            .format(data.dtype))
    else:
350
        raise TypeError("Unknown type({})".format(type(data).__name__))
351
    return (ptr_data, type_data, data)  # return `data` to avoid the temporary copy is freed
wxchan's avatar
wxchan committed
352

wxchan's avatar
wxchan committed
353

354
355
356
357
358
359
360
361
362
363
364
def _get_bad_pandas_dtypes(dtypes):
    pandas_dtype_mapper = {'int8': 'int', 'int16': 'int', 'int32': 'int',
                           'int64': 'int', 'uint8': 'int', 'uint16': 'int',
                           'uint32': 'int', 'uint64': 'int', 'bool': 'int',
                           'float16': 'float', 'float32': 'float', 'float64': 'float'}
    bad_indices = [i for i, dtype in enumerate(dtypes) if (dtype.name not in pandas_dtype_mapper
                                                           and (not is_dtype_sparse(dtype)
                                                                or dtype.subtype.name not in pandas_dtype_mapper))]
    return bad_indices


365
def _data_from_pandas(data, feature_name, categorical_feature, pandas_categorical):
366
    if isinstance(data, DataFrame):
367
368
        if len(data.shape) != 2 or data.shape[0] < 1:
            raise ValueError('Input data must be 2 dimensional and non empty.')
369
370
        if feature_name == 'auto' or feature_name is None:
            data = data.rename(columns=str)
371
372
        cat_cols = list(data.select_dtypes(include=['category']).columns)
        cat_cols_not_ordered = [col for col in cat_cols if not data[col].cat.ordered]
373
374
375
376
377
        if pandas_categorical is None:  # train dataset
            pandas_categorical = [list(data[col].cat.categories) for col in cat_cols]
        else:
            if len(cat_cols) != len(pandas_categorical):
                raise ValueError('train and valid dataset categorical_feature do not match.')
378
            for col, category in zip_(cat_cols, pandas_categorical):
379
380
                if list(data[col].cat.categories) != list(category):
                    data[col] = data[col].cat.set_categories(category)
381
        if len(cat_cols):  # cat_cols is list
382
            data = data.copy()  # not alter origin DataFrame
383
            data[cat_cols] = data[cat_cols].apply(lambda x: x.cat.codes).replace({-1: np.nan})
384
385
386
        if categorical_feature is not None:
            if feature_name is None:
                feature_name = list(data.columns)
387
            if categorical_feature == 'auto':  # use cat cols from DataFrame
388
                categorical_feature = cat_cols_not_ordered
389
390
            else:  # use cat cols specified by user
                categorical_feature = list(categorical_feature)
391
392
        if feature_name == 'auto':
            feature_name = list(data.columns)
393
394
        bad_indices = _get_bad_pandas_dtypes(data.dtypes)
        if bad_indices:
395
            raise ValueError("DataFrame.dtypes for data must be int, float or bool.\n"
396
                             "Did not expect the data types in the following fields: "
397
                             + ', '.join(data.columns[bad_indices]))
398
399
400
        data = data.values
        if data.dtype != np.float32 and data.dtype != np.float64:
            data = data.astype(np.float32)
401
402
403
404
405
406
    else:
        if feature_name == 'auto':
            feature_name = None
        if categorical_feature == 'auto':
            categorical_feature = None
    return data, feature_name, categorical_feature, pandas_categorical
407
408
409
410
411
412


def _label_from_pandas(label):
    if isinstance(label, DataFrame):
        if len(label.columns) > 1:
            raise ValueError('DataFrame for label cannot have multiple columns')
413
        if _get_bad_pandas_dtypes(label.dtypes):
414
            raise ValueError('DataFrame.dtypes for label must be int, float or bool')
415
        label = np.ravel(label.values.astype(np.float32, copy=False))
416
417
418
    return label


419
420
421
422
423
424
425
426
427
428
429
def _dump_pandas_categorical(pandas_categorical, file_name=None):
    pandas_str = ('\npandas_categorical:'
                  + json.dumps(pandas_categorical, default=json_default_with_numpy)
                  + '\n')
    if file_name is not None:
        with open(file_name, 'a') as f:
            f.write(pandas_str)
    return pandas_str


def _load_pandas_categorical(file_name=None, model_str=None):
430
431
    pandas_key = 'pandas_categorical:'
    offset = -len(pandas_key)
432
    if file_name is not None:
433
434
435
436
437
438
439
440
441
442
443
444
445
        max_offset = -os.path.getsize(file_name)
        with open(file_name, 'rb') as f:
            while True:
                if offset < max_offset:
                    offset = max_offset
                f.seek(offset, os.SEEK_END)
                lines = f.readlines()
                if len(lines) >= 2:
                    break
                offset *= 2
        last_line = decode_string(lines[-1]).strip()
        if not last_line.startswith(pandas_key):
            last_line = decode_string(lines[-2]).strip()
446
    elif model_str is not None:
447
448
449
450
451
452
        idx = model_str.rfind('\n', 0, offset)
        last_line = model_str[idx:].strip()
    if last_line.startswith(pandas_key):
        return json.loads(last_line[len(pandas_key):])
    else:
        return None
453
454


Guolin Ke's avatar
Guolin Ke committed
455
class _InnerPredictor(object):
456
457
458
459
460
    """_InnerPredictor of LightGBM.

    Not exposed to user.
    Used only for prediction, usually used for continued training.

Nikita Titov's avatar
Nikita Titov committed
461
462
463
    .. note::

        Can be converted from Booster, but cannot be converted to Booster.
Guolin Ke's avatar
Guolin Ke committed
464
    """
465

466
    def __init__(self, model_file=None, booster_handle=None, pred_parameter=None):
467
        """Initialize the _InnerPredictor.
wxchan's avatar
wxchan committed
468
469
470

        Parameters
        ----------
471
        model_file : string or None, optional (default=None)
wxchan's avatar
wxchan committed
472
            Path to the model file.
473
474
475
476
        booster_handle : object or None, optional (default=None)
            Handle of Booster.
        pred_parameter: dict or None, optional (default=None)
            Other parameters for the prediciton.
wxchan's avatar
wxchan committed
477
478
479
480
481
        """
        self.handle = ctypes.c_void_p()
        self.__is_manage_handle = True
        if model_file is not None:
            """Prediction task"""
Guolin Ke's avatar
Guolin Ke committed
482
            out_num_iterations = ctypes.c_int(0)
wxchan's avatar
wxchan committed
483
484
485
486
            _safe_call(_LIB.LGBM_BoosterCreateFromModelfile(
                c_str(model_file),
                ctypes.byref(out_num_iterations),
                ctypes.byref(self.handle)))
Guolin Ke's avatar
Guolin Ke committed
487
            out_num_class = ctypes.c_int(0)
wxchan's avatar
wxchan committed
488
489
490
491
            _safe_call(_LIB.LGBM_BoosterGetNumClasses(
                self.handle,
                ctypes.byref(out_num_class)))
            self.num_class = out_num_class.value
492
            self.num_total_iteration = out_num_iterations.value
493
            self.pandas_categorical = _load_pandas_categorical(file_name=model_file)
wxchan's avatar
wxchan committed
494
        elif booster_handle is not None:
Guolin Ke's avatar
Guolin Ke committed
495
            self.__is_manage_handle = False
wxchan's avatar
wxchan committed
496
            self.handle = booster_handle
Guolin Ke's avatar
Guolin Ke committed
497
            out_num_class = ctypes.c_int(0)
wxchan's avatar
wxchan committed
498
499
500
501
            _safe_call(_LIB.LGBM_BoosterGetNumClasses(
                self.handle,
                ctypes.byref(out_num_class)))
            self.num_class = out_num_class.value
502
            self.num_total_iteration = self.current_iteration()
503
            self.pandas_categorical = None
wxchan's avatar
wxchan committed
504
        else:
505
            raise TypeError('Need model_file or booster_handle to create a predictor')
wxchan's avatar
wxchan committed
506

507
508
        pred_parameter = {} if pred_parameter is None else pred_parameter
        self.pred_parameter = param_dict_to_str(pred_parameter)
cbecker's avatar
cbecker committed
509

wxchan's avatar
wxchan committed
510
    def __del__(self):
511
512
513
514
515
        try:
            if self.__is_manage_handle:
                _safe_call(_LIB.LGBM_BoosterFree(self.handle))
        except AttributeError:
            pass
wxchan's avatar
wxchan committed
516

517
518
519
520
521
    def __getstate__(self):
        this = self.__dict__.copy()
        this.pop('handle', None)
        return this

522
    def predict(self, data, start_iteration=0, num_iteration=-1,
523
                raw_score=False, pred_leaf=False, pred_contrib=False, data_has_header=False,
wxchan's avatar
wxchan committed
524
                is_reshape=True):
525
        """Predict logic.
wxchan's avatar
wxchan committed
526
527
528

        Parameters
        ----------
529
        data : string, numpy array, pandas DataFrame, H2O DataTable's Frame or scipy.sparse
530
531
            Data source for prediction.
            When data type is string, it represents the path of txt file.
532
533
        start_iteration : int, optional (default=0)
            Start index of the iteration to predict.
534
535
536
537
538
539
540
541
542
543
544
545
546
        num_iteration : int, optional (default=-1)
            Iteration used for prediction.
        raw_score : bool, optional (default=False)
            Whether to predict raw scores.
        pred_leaf : bool, optional (default=False)
            Whether to predict leaf index.
        pred_contrib : bool, optional (default=False)
            Whether to predict feature contributions.
        data_has_header : bool, optional (default=False)
            Whether data has header.
            Used only for txt data.
        is_reshape : bool, optional (default=True)
            Whether to reshape to (nrow, ncol).
wxchan's avatar
wxchan committed
547
548
549

        Returns
        -------
550
        result : numpy array, scipy.sparse or list of scipy.sparse
551
            Prediction result.
552
            Can be sparse or a list of sparse objects (each element represents predictions for one class) for feature contributions (when ``pred_contrib=True``).
wxchan's avatar
wxchan committed
553
        """
wxchan's avatar
wxchan committed
554
        if isinstance(data, Dataset):
555
            raise TypeError("Cannot use Dataset instance for prediction, please use raw data instead")
556
        data = _data_from_pandas(data, None, None, self.pandas_categorical)[0]
wxchan's avatar
wxchan committed
557
558
559
560
561
        predict_type = C_API_PREDICT_NORMAL
        if raw_score:
            predict_type = C_API_PREDICT_RAW_SCORE
        if pred_leaf:
            predict_type = C_API_PREDICT_LEAF_INDEX
562
563
        if pred_contrib:
            predict_type = C_API_PREDICT_CONTRIB
wxchan's avatar
wxchan committed
564
        int_data_has_header = 1 if data_has_header else 0
cbecker's avatar
cbecker committed
565

wxchan's avatar
wxchan committed
566
        if isinstance(data, string_type):
567
            with _TempFile() as f:
wxchan's avatar
wxchan committed
568
569
570
                _safe_call(_LIB.LGBM_BoosterPredictForFile(
                    self.handle,
                    c_str(data),
Guolin Ke's avatar
Guolin Ke committed
571
572
                    ctypes.c_int(int_data_has_header),
                    ctypes.c_int(predict_type),
573
                    ctypes.c_int(start_iteration),
Guolin Ke's avatar
Guolin Ke committed
574
                    ctypes.c_int(num_iteration),
575
                    c_str(self.pred_parameter),
wxchan's avatar
wxchan committed
576
577
                    c_str(f.name)))
                lines = f.readlines()
578
579
                nrow = len(lines)
                preds = [float(token) for line in lines for token in line.split('\t')]
Guolin Ke's avatar
Guolin Ke committed
580
                preds = np.array(preds, dtype=np.float64, copy=False)
wxchan's avatar
wxchan committed
581
        elif isinstance(data, scipy.sparse.csr_matrix):
582
            preds, nrow = self.__pred_for_csr(data, start_iteration, num_iteration, predict_type)
Guolin Ke's avatar
Guolin Ke committed
583
        elif isinstance(data, scipy.sparse.csc_matrix):
584
            preds, nrow = self.__pred_for_csc(data, start_iteration, num_iteration, predict_type)
wxchan's avatar
wxchan committed
585
        elif isinstance(data, np.ndarray):
586
            preds, nrow = self.__pred_for_np2d(data, start_iteration, num_iteration, predict_type)
587
588
589
        elif isinstance(data, list):
            try:
                data = np.array(data)
590
            except BaseException:
591
                raise ValueError('Cannot convert data list to numpy array.')
592
            preds, nrow = self.__pred_for_np2d(data, start_iteration, num_iteration, predict_type)
593
        elif isinstance(data, DataTable):
594
            preds, nrow = self.__pred_for_np2d(data.to_numpy(), start_iteration, num_iteration, predict_type)
wxchan's avatar
wxchan committed
595
596
        else:
            try:
597
                warnings.warn('Converting data to scipy sparse matrix.')
wxchan's avatar
wxchan committed
598
                csr = scipy.sparse.csr_matrix(data)
599
            except BaseException:
600
                raise TypeError('Cannot predict data for type {}'.format(type(data).__name__))
601
            preds, nrow = self.__pred_for_csr(csr, start_iteration, num_iteration, predict_type)
wxchan's avatar
wxchan committed
602
603
        if pred_leaf:
            preds = preds.astype(np.int32)
604
605
        is_sparse = scipy.sparse.issparse(preds) or isinstance(preds, list)
        if is_reshape and not is_sparse and preds.size != nrow:
wxchan's avatar
wxchan committed
606
            if preds.size % nrow == 0:
607
                preds = preds.reshape(nrow, -1)
wxchan's avatar
wxchan committed
608
            else:
609
                raise ValueError('Length of predict result (%d) cannot be divide nrow (%d)'
wxchan's avatar
wxchan committed
610
611
612
                                 % (preds.size, nrow))
        return preds

613
    def __get_num_preds(self, start_iteration, num_iteration, nrow, predict_type):
614
        """Get size of prediction result."""
615
616
617
618
619
        if nrow > MAX_INT32:
            raise LightGBMError('LightGBM cannot perform prediction for data'
                                'with number of rows greater than MAX_INT32 (%d).\n'
                                'You can split your data into chunks'
                                'and then concatenate predictions for them' % MAX_INT32)
Guolin Ke's avatar
Guolin Ke committed
620
621
622
        n_preds = ctypes.c_int64(0)
        _safe_call(_LIB.LGBM_BoosterCalcNumPredict(
            self.handle,
Guolin Ke's avatar
Guolin Ke committed
623
624
            ctypes.c_int(nrow),
            ctypes.c_int(predict_type),
625
            ctypes.c_int(start_iteration),
Guolin Ke's avatar
Guolin Ke committed
626
            ctypes.c_int(num_iteration),
Guolin Ke's avatar
Guolin Ke committed
627
628
            ctypes.byref(n_preds)))
        return n_preds.value
wxchan's avatar
wxchan committed
629

630
    def __pred_for_np2d(self, mat, start_iteration, num_iteration, predict_type):
631
        """Predict for a 2-D numpy matrix."""
wxchan's avatar
wxchan committed
632
        if len(mat.shape) != 2:
633
            raise ValueError('Input numpy.ndarray or list must be 2 dimensional')
wxchan's avatar
wxchan committed
634

635
        def inner_predict(mat, start_iteration, num_iteration, predict_type, preds=None):
636
637
            if mat.dtype == np.float32 or mat.dtype == np.float64:
                data = np.array(mat.reshape(mat.size), dtype=mat.dtype, copy=False)
638
            else:  # change non-float data to float data, need to copy
639
640
                data = np.array(mat.reshape(mat.size), dtype=np.float32)
            ptr_data, type_ptr_data, _ = c_float_array(data)
641
            n_preds = self.__get_num_preds(start_iteration, num_iteration, mat.shape[0], predict_type)
642
643
644
645
646
647
648
649
650
651
652
653
654
            if preds is None:
                preds = np.zeros(n_preds, dtype=np.float64)
            elif len(preds.shape) != 1 or len(preds) != n_preds:
                raise ValueError("Wrong length of pre-allocated predict array")
            out_num_preds = ctypes.c_int64(0)
            _safe_call(_LIB.LGBM_BoosterPredictForMat(
                self.handle,
                ptr_data,
                ctypes.c_int(type_ptr_data),
                ctypes.c_int(mat.shape[0]),
                ctypes.c_int(mat.shape[1]),
                ctypes.c_int(C_API_IS_ROW_MAJOR),
                ctypes.c_int(predict_type),
655
                ctypes.c_int(start_iteration),
656
657
658
659
660
661
662
663
664
665
666
667
                ctypes.c_int(num_iteration),
                c_str(self.pred_parameter),
                ctypes.byref(out_num_preds),
                preds.ctypes.data_as(ctypes.POINTER(ctypes.c_double))))
            if n_preds != out_num_preds.value:
                raise ValueError("Wrong length for predict results")
            return preds, mat.shape[0]

        nrow = mat.shape[0]
        if nrow > MAX_INT32:
            sections = np.arange(start=MAX_INT32, stop=nrow, step=MAX_INT32)
            # __get_num_preds() cannot work with nrow > MAX_INT32, so calculate overall number of predictions piecemeal
668
            n_preds = [self.__get_num_preds(start_iteration, num_iteration, i, predict_type) for i in np.diff([0] + list(sections) + [nrow])]
669
670
            n_preds_sections = np.array([0] + n_preds, dtype=np.intp).cumsum()
            preds = np.zeros(sum(n_preds), dtype=np.float64)
671
672
            for chunk, (start_idx_pred, end_idx_pred) in zip_(np.array_split(mat, sections),
                                                              zip_(n_preds_sections, n_preds_sections[1:])):
673
                # avoid memory consumption by arrays concatenation operations
674
                inner_predict(chunk, start_iteration, num_iteration, predict_type, preds[start_idx_pred:end_idx_pred])
675
            return preds, nrow
wxchan's avatar
wxchan committed
676
        else:
677
            return inner_predict(mat, start_iteration, num_iteration, predict_type)
wxchan's avatar
wxchan committed
678

679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
    def __create_sparse_native(self, cs, out_shape, out_ptr_indptr, out_ptr_indices, out_ptr_data,
                               indptr_type, data_type, is_csr=True):
        # create numpy array from output arrays
        data_indices_len = out_shape[0]
        indptr_len = out_shape[1]
        if indptr_type == C_API_DTYPE_INT32:
            out_indptr = cint32_array_to_numpy(out_ptr_indptr, indptr_len)
        elif indptr_type == C_API_DTYPE_INT64:
            out_indptr = cint64_array_to_numpy(out_ptr_indptr, indptr_len)
        else:
            raise TypeError("Expected int32 or int64 type for indptr")
        if data_type == C_API_DTYPE_FLOAT32:
            out_data = cfloat32_array_to_numpy(out_ptr_data, data_indices_len)
        elif data_type == C_API_DTYPE_FLOAT64:
            out_data = cfloat64_array_to_numpy(out_ptr_data, data_indices_len)
        else:
            raise TypeError("Expected float32 or float64 type for data")
        out_indices = cint32_array_to_numpy(out_ptr_indices, data_indices_len)
        # break up indptr based on number of rows (note more than one matrix in multiclass case)
        per_class_indptr_shape = cs.indptr.shape[0]
        # for CSC there is extra column added
        if not is_csr:
            per_class_indptr_shape += 1
        out_indptr_arrays = np.split(out_indptr, out_indptr.shape[0] / per_class_indptr_shape)
        # reformat output into a csr or csc matrix or list of csr or csc matrices
        cs_output_matrices = []
        offset = 0
        for cs_indptr in out_indptr_arrays:
            matrix_indptr_len = cs_indptr[cs_indptr.shape[0] - 1]
            cs_indices = out_indices[offset + cs_indptr[0]:offset + matrix_indptr_len]
            cs_data = out_data[offset + cs_indptr[0]:offset + matrix_indptr_len]
            offset += matrix_indptr_len
            # same shape as input csr or csc matrix except extra column for expected value
            cs_shape = [cs.shape[0], cs.shape[1] + 1]
            # note: make sure we copy data as it will be deallocated next
            if is_csr:
                cs_output_matrices.append(scipy.sparse.csr_matrix((cs_data, cs_indices, cs_indptr), cs_shape))
            else:
                cs_output_matrices.append(scipy.sparse.csc_matrix((cs_data, cs_indices, cs_indptr), cs_shape))
        # free the temporary native indptr, indices, and data
        _safe_call(_LIB.LGBM_BoosterFreePredictSparse(out_ptr_indptr, out_ptr_indices, out_ptr_data,
                                                      ctypes.c_int(indptr_type), ctypes.c_int(data_type)))
        if len(cs_output_matrices) == 1:
            return cs_output_matrices[0]
        return cs_output_matrices

725
    def __pred_for_csr(self, csr, start_iteration, num_iteration, predict_type):
726
        """Predict for a CSR data."""
727
        def inner_predict(csr, start_iteration, num_iteration, predict_type, preds=None):
728
            nrow = len(csr.indptr) - 1
729
            n_preds = self.__get_num_preds(start_iteration, num_iteration, nrow, predict_type)
730
731
732
733
734
735
736
737
738
            if preds is None:
                preds = np.zeros(n_preds, dtype=np.float64)
            elif len(preds.shape) != 1 or len(preds) != n_preds:
                raise ValueError("Wrong length of pre-allocated predict array")
            out_num_preds = ctypes.c_int64(0)

            ptr_indptr, type_ptr_indptr, __ = c_int_array(csr.indptr)
            ptr_data, type_ptr_data, _ = c_float_array(csr.data)

739
            assert csr.shape[1] <= MAX_INT32
740
            csr_indices = csr.indices.astype(np.int32, copy=False)
741

742
743
744
745
            _safe_call(_LIB.LGBM_BoosterPredictForCSR(
                self.handle,
                ptr_indptr,
                ctypes.c_int32(type_ptr_indptr),
746
                csr_indices.ctypes.data_as(ctypes.POINTER(ctypes.c_int32)),
747
748
749
750
751
752
                ptr_data,
                ctypes.c_int(type_ptr_data),
                ctypes.c_int64(len(csr.indptr)),
                ctypes.c_int64(len(csr.data)),
                ctypes.c_int64(csr.shape[1]),
                ctypes.c_int(predict_type),
753
                ctypes.c_int(start_iteration),
754
755
756
757
758
759
760
                ctypes.c_int(num_iteration),
                c_str(self.pred_parameter),
                ctypes.byref(out_num_preds),
                preds.ctypes.data_as(ctypes.POINTER(ctypes.c_double))))
            if n_preds != out_num_preds.value:
                raise ValueError("Wrong length for predict results")
            return preds, nrow
wxchan's avatar
wxchan committed
761

762
        def inner_predict_sparse(csr, start_iteration, num_iteration, predict_type):
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
            ptr_indptr, type_ptr_indptr, __ = c_int_array(csr.indptr)
            ptr_data, type_ptr_data, _ = c_float_array(csr.data)
            csr_indices = csr.indices.astype(np.int32, copy=False)
            matrix_type = C_API_MATRIX_TYPE_CSR
            if type_ptr_indptr == C_API_DTYPE_INT32:
                out_ptr_indptr = ctypes.POINTER(ctypes.c_int32)()
            else:
                out_ptr_indptr = ctypes.POINTER(ctypes.c_int64)()
            out_ptr_indices = ctypes.POINTER(ctypes.c_int32)()
            if type_ptr_data == C_API_DTYPE_FLOAT32:
                out_ptr_data = ctypes.POINTER(ctypes.c_float)()
            else:
                out_ptr_data = ctypes.POINTER(ctypes.c_double)()
            out_shape = np.zeros(2, dtype=np.int64)
            _safe_call(_LIB.LGBM_BoosterPredictSparseOutput(
                self.handle,
                ptr_indptr,
                ctypes.c_int32(type_ptr_indptr),
                csr_indices.ctypes.data_as(ctypes.POINTER(ctypes.c_int32)),
                ptr_data,
                ctypes.c_int(type_ptr_data),
                ctypes.c_int64(len(csr.indptr)),
                ctypes.c_int64(len(csr.data)),
                ctypes.c_int64(csr.shape[1]),
                ctypes.c_int(predict_type),
788
                ctypes.c_int(start_iteration),
789
790
791
792
793
794
795
796
797
798
799
800
801
                ctypes.c_int(num_iteration),
                c_str(self.pred_parameter),
                ctypes.c_int(matrix_type),
                out_shape.ctypes.data_as(ctypes.POINTER(ctypes.c_int64)),
                ctypes.byref(out_ptr_indptr),
                ctypes.byref(out_ptr_indices),
                ctypes.byref(out_ptr_data)))
            matrices = self.__create_sparse_native(csr, out_shape, out_ptr_indptr, out_ptr_indices, out_ptr_data,
                                                   type_ptr_indptr, type_ptr_data, is_csr=True)
            nrow = len(csr.indptr) - 1
            return matrices, nrow

        if predict_type == C_API_PREDICT_CONTRIB:
802
            return inner_predict_sparse(csr, start_iteration, num_iteration, predict_type)
803
804
805
806
        nrow = len(csr.indptr) - 1
        if nrow > MAX_INT32:
            sections = [0] + list(np.arange(start=MAX_INT32, stop=nrow, step=MAX_INT32)) + [nrow]
            # __get_num_preds() cannot work with nrow > MAX_INT32, so calculate overall number of predictions piecemeal
807
            n_preds = [self.__get_num_preds(start_iteration, num_iteration, i, predict_type) for i in np.diff(sections)]
808
809
810
811
812
            n_preds_sections = np.array([0] + n_preds, dtype=np.intp).cumsum()
            preds = np.zeros(sum(n_preds), dtype=np.float64)
            for (start_idx, end_idx), (start_idx_pred, end_idx_pred) in zip_(zip_(sections, sections[1:]),
                                                                             zip_(n_preds_sections, n_preds_sections[1:])):
                # avoid memory consumption by arrays concatenation operations
813
                inner_predict(csr[start_idx:end_idx], start_iteration, num_iteration, predict_type, preds[start_idx_pred:end_idx_pred])
814
815
            return preds, nrow
        else:
816
            return inner_predict(csr, start_iteration, num_iteration, predict_type)
Guolin Ke's avatar
Guolin Ke committed
817

818
    def __pred_for_csc(self, csc, start_iteration, num_iteration, predict_type):
819
        """Predict for a CSC data."""
820
        def inner_predict_sparse(csc, start_iteration, num_iteration, predict_type):
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
            ptr_indptr, type_ptr_indptr, __ = c_int_array(csc.indptr)
            ptr_data, type_ptr_data, _ = c_float_array(csc.data)
            csc_indices = csc.indices.astype(np.int32, copy=False)
            matrix_type = C_API_MATRIX_TYPE_CSC
            if type_ptr_indptr == C_API_DTYPE_INT32:
                out_ptr_indptr = ctypes.POINTER(ctypes.c_int32)()
            else:
                out_ptr_indptr = ctypes.POINTER(ctypes.c_int64)()
            out_ptr_indices = ctypes.POINTER(ctypes.c_int32)()
            if type_ptr_data == C_API_DTYPE_FLOAT32:
                out_ptr_data = ctypes.POINTER(ctypes.c_float)()
            else:
                out_ptr_data = ctypes.POINTER(ctypes.c_double)()
            out_shape = np.zeros(2, dtype=np.int64)
            _safe_call(_LIB.LGBM_BoosterPredictSparseOutput(
                self.handle,
                ptr_indptr,
                ctypes.c_int32(type_ptr_indptr),
                csc_indices.ctypes.data_as(ctypes.POINTER(ctypes.c_int32)),
                ptr_data,
                ctypes.c_int(type_ptr_data),
                ctypes.c_int64(len(csc.indptr)),
                ctypes.c_int64(len(csc.data)),
                ctypes.c_int64(csc.shape[0]),
                ctypes.c_int(predict_type),
846
                ctypes.c_int(start_iteration),
847
848
849
850
851
852
853
854
855
856
857
858
                ctypes.c_int(num_iteration),
                c_str(self.pred_parameter),
                ctypes.c_int(matrix_type),
                out_shape.ctypes.data_as(ctypes.POINTER(ctypes.c_int64)),
                ctypes.byref(out_ptr_indptr),
                ctypes.byref(out_ptr_indices),
                ctypes.byref(out_ptr_data)))
            matrices = self.__create_sparse_native(csc, out_shape, out_ptr_indptr, out_ptr_indices, out_ptr_data,
                                                   type_ptr_indptr, type_ptr_data, is_csr=False)
            nrow = csc.shape[0]
            return matrices, nrow

Guolin Ke's avatar
Guolin Ke committed
859
        nrow = csc.shape[0]
860
        if nrow > MAX_INT32:
861
            return self.__pred_for_csr(csc.tocsr(), start_iteration, num_iteration, predict_type)
862
        if predict_type == C_API_PREDICT_CONTRIB:
863
864
            return inner_predict_sparse(csc, start_iteration, num_iteration, predict_type)
        n_preds = self.__get_num_preds(start_iteration, num_iteration, nrow, predict_type)
Guolin Ke's avatar
Guolin Ke committed
865
866
867
        preds = np.zeros(n_preds, dtype=np.float64)
        out_num_preds = ctypes.c_int64(0)

868
869
        ptr_indptr, type_ptr_indptr, __ = c_int_array(csc.indptr)
        ptr_data, type_ptr_data, _ = c_float_array(csc.data)
Guolin Ke's avatar
Guolin Ke committed
870

871
        assert csc.shape[0] <= MAX_INT32
872
        csc_indices = csc.indices.astype(np.int32, copy=False)
873

Guolin Ke's avatar
Guolin Ke committed
874
875
876
        _safe_call(_LIB.LGBM_BoosterPredictForCSC(
            self.handle,
            ptr_indptr,
Guolin Ke's avatar
Guolin Ke committed
877
            ctypes.c_int32(type_ptr_indptr),
878
            csc_indices.ctypes.data_as(ctypes.POINTER(ctypes.c_int32)),
Guolin Ke's avatar
Guolin Ke committed
879
            ptr_data,
Guolin Ke's avatar
Guolin Ke committed
880
881
882
883
884
            ctypes.c_int(type_ptr_data),
            ctypes.c_int64(len(csc.indptr)),
            ctypes.c_int64(len(csc.data)),
            ctypes.c_int64(csc.shape[0]),
            ctypes.c_int(predict_type),
885
            ctypes.c_int(start_iteration),
Guolin Ke's avatar
Guolin Ke committed
886
            ctypes.c_int(num_iteration),
887
            c_str(self.pred_parameter),
Guolin Ke's avatar
Guolin Ke committed
888
            ctypes.byref(out_num_preds),
wxchan's avatar
wxchan committed
889
            preds.ctypes.data_as(ctypes.POINTER(ctypes.c_double))))
wxchan's avatar
wxchan committed
890
        if n_preds != out_num_preds.value:
891
            raise ValueError("Wrong length for predict results")
wxchan's avatar
wxchan committed
892
893
        return preds, nrow

894
895
896
897
898
899
900
901
902
903
904
905
906
907
    def current_iteration(self):
        """Get the index of the current iteration.

        Returns
        -------
        cur_iter : int
            The index of the current iteration.
        """
        out_cur_iter = ctypes.c_int(0)
        _safe_call(_LIB.LGBM_BoosterGetCurrentIteration(
            self.handle,
            ctypes.byref(out_cur_iter)))
        return out_cur_iter.value

wxchan's avatar
wxchan committed
908

wxchan's avatar
wxchan committed
909
910
class Dataset(object):
    """Dataset in LightGBM."""
911

912
    def __init__(self, data, label=None, reference=None,
913
                 weight=None, group=None, init_score=None, silent=False,
914
                 feature_name='auto', categorical_feature='auto', params=None,
wxchan's avatar
wxchan committed
915
                 free_raw_data=True):
916
        """Initialize Dataset.
917

wxchan's avatar
wxchan committed
918
919
        Parameters
        ----------
920
        data : string, numpy array, pandas DataFrame, H2O DataTable's Frame, scipy.sparse or list of numpy arrays
wxchan's avatar
wxchan committed
921
            Data source of Dataset.
922
            If string, it represents the path to txt file.
923
        label : list, numpy 1-D array, pandas Series / one-column DataFrame or None, optional (default=None)
924
925
926
            Label of the data.
        reference : Dataset or None, optional (default=None)
            If this is Dataset for validation, training data should be used as reference.
927
        weight : list, numpy 1-D array, pandas Series or None, optional (default=None)
wxchan's avatar
wxchan committed
928
            Weight for each instance.
929
        group : list, numpy 1-D array, pandas Series or None, optional (default=None)
930
            Group/query size for Dataset.
931
        init_score : list, numpy 1-D array, pandas Series or None, optional (default=None)
932
            Init score for Dataset.
933
934
935
936
937
938
939
940
941
        silent : bool, optional (default=False)
            Whether to print messages during construction.
        feature_name : list of strings or 'auto', optional (default="auto")
            Feature names.
            If 'auto' and data is pandas DataFrame, data columns names are used.
        categorical_feature : list of strings or int, or 'auto', optional (default="auto")
            Categorical features.
            If list of int, interpreted as indices.
            If list of strings, interpreted as feature names (need to specify ``feature_name`` as well).
942
            If 'auto' and data is pandas DataFrame, pandas unordered categorical columns are used.
943
            All values in categorical features should be less than int32 max value (2147483647).
944
            Large values could be memory consuming. Consider using consecutive integers starting from zero.
945
            All negative values in categorical features will be treated as missing values.
946
            The output cannot be monotonically constrained with respect to a categorical feature.
Nikita Titov's avatar
Nikita Titov committed
947
        params : dict or None, optional (default=None)
948
            Other parameters for Dataset.
Nikita Titov's avatar
Nikita Titov committed
949
        free_raw_data : bool, optional (default=True)
950
            If True, raw data is freed after constructing inner Dataset.
wxchan's avatar
wxchan committed
951
        """
wxchan's avatar
wxchan committed
952
953
954
955
956
957
        self.handle = None
        self.data = data
        self.label = label
        self.reference = reference
        self.weight = weight
        self.group = group
958
        self.init_score = init_score
wxchan's avatar
wxchan committed
959
960
        self.silent = silent
        self.feature_name = feature_name
961
        self.categorical_feature = categorical_feature
962
        self.params = copy.deepcopy(params)
wxchan's avatar
wxchan committed
963
964
        self.free_raw_data = free_raw_data
        self.used_indices = None
965
        self.need_slice = True
wxchan's avatar
wxchan committed
966
        self._predictor = None
967
        self.pandas_categorical = None
968
        self.params_back_up = None
969
970
        self.feature_penalty = None
        self.monotone_constraints = None
971
        self.version = 0
wxchan's avatar
wxchan committed
972
973

    def __del__(self):
974
975
976
977
        try:
            self._free_handle()
        except AttributeError:
            pass
978

979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
    def get_params(self):
        """Get the used parameters in the Dataset.

        Returns
        -------
        params : dict or None
            The used parameters in this Dataset object.
        """
        if self.params is not None:
            # no min_data, nthreads and verbose in this function
            dataset_params = _ConfigAliases.get("bin_construct_sample_cnt",
                                                "categorical_feature",
                                                "data_random_seed",
                                                "enable_bundle",
                                                "feature_pre_filter",
                                                "forcedbins_filename",
                                                "group_column",
                                                "header",
                                                "ignore_column",
                                                "is_enable_sparse",
                                                "label_column",
                                                "max_bin",
                                                "max_bin_by_feature",
                                                "min_data_in_bin",
                                                "pre_partition",
                                                "two_round",
                                                "use_missing",
                                                "weight_column",
                                                "zero_as_missing")
            return {k: v for k, v in self.params.items() if k in dataset_params}

1010
    def _free_handle(self):
1011
        if self.handle is not None:
1012
            _safe_call(_LIB.LGBM_DatasetFree(self.handle))
1013
            self.handle = None
Guolin Ke's avatar
Guolin Ke committed
1014
1015
1016
        self.need_slice = True
        if self.used_indices is not None:
            self.data = None
Nikita Titov's avatar
Nikita Titov committed
1017
        return self
wxchan's avatar
wxchan committed
1018

Guolin Ke's avatar
Guolin Ke committed
1019
1020
1021
1022
    def _set_init_score_by_predictor(self, predictor, data, used_indices=None):
        data_has_header = False
        if isinstance(data, string_type):
            # check data has header or not
1023
            data_has_header = any(self.params.get(alias, False) for alias in _ConfigAliases.get("header"))
Guolin Ke's avatar
Guolin Ke committed
1024
        num_data = self.num_data()
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
        if predictor is not None:
            init_score = predictor.predict(data,
                                           raw_score=True,
                                           data_has_header=data_has_header,
                                           is_reshape=False)
            if used_indices is not None:
                assert not self.need_slice
                if isinstance(data, string_type):
                    sub_init_score = np.zeros(num_data * predictor.num_class, dtype=np.float32)
                    assert num_data == len(used_indices)
                    for i in range_(len(used_indices)):
                        for j in range_(predictor.num_class):
                            sub_init_score[i * predictor.num_class + j] = init_score[used_indices[i] * predictor.num_class + j]
                    init_score = sub_init_score
            if predictor.num_class > 1:
                # need to regroup init_score
                new_init_score = np.zeros(init_score.size, dtype=np.float32)
                for i in range_(num_data):
Guolin Ke's avatar
Guolin Ke committed
1043
                    for j in range_(predictor.num_class):
1044
1045
1046
1047
1048
1049
                        new_init_score[j * num_data + i] = init_score[i * predictor.num_class + j]
                init_score = new_init_score
        elif self.init_score is not None:
            init_score = np.zeros(self.init_score.shape, dtype=np.float32)
        else:
            return self
Guolin Ke's avatar
Guolin Ke committed
1050
1051
        self.set_init_score(init_score)

1052
    def _lazy_init(self, data, label=None, reference=None,
1053
                   weight=None, group=None, init_score=None, predictor=None,
wxchan's avatar
wxchan committed
1054
                   silent=False, feature_name='auto',
1055
                   categorical_feature='auto', params=None):
wxchan's avatar
wxchan committed
1056
1057
        if data is None:
            self.handle = None
Nikita Titov's avatar
Nikita Titov committed
1058
            return self
Guolin Ke's avatar
Guolin Ke committed
1059
1060
1061
        if reference is not None:
            self.pandas_categorical = reference.pandas_categorical
            categorical_feature = reference.categorical_feature
1062
1063
1064
1065
        data, feature_name, categorical_feature, self.pandas_categorical = _data_from_pandas(data,
                                                                                             feature_name,
                                                                                             categorical_feature,
                                                                                             self.pandas_categorical)
wxchan's avatar
wxchan committed
1066
        label = _label_from_pandas(label)
Guolin Ke's avatar
Guolin Ke committed
1067

1068
        # process for args
wxchan's avatar
wxchan committed
1069
        params = {} if params is None else params
1070
1071
1072
        args_names = (getattr(self.__class__, '_lazy_init')
                      .__code__
                      .co_varnames[:getattr(self.__class__, '_lazy_init').__code__.co_argcount])
1073
1074
        for key, _ in params.items():
            if key in args_names:
1075
1076
1077
                warnings.warn('{0} keyword has been found in `params` and will be ignored.\n'
                              'Please use {0} argument of the Dataset constructor to pass this parameter.'
                              .format(key))
1078
        # user can set verbose with params, it has higher priority
1079
        if not any(verbose_alias in params for verbose_alias in _ConfigAliases.get("verbosity")) and silent:
1080
            params["verbose"] = -1
1081
        # get categorical features
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
        if categorical_feature is not None:
            categorical_indices = set()
            feature_dict = {}
            if feature_name is not None:
                feature_dict = {name: i for i, name in enumerate(feature_name)}
            for name in categorical_feature:
                if isinstance(name, string_type) and name in feature_dict:
                    categorical_indices.add(feature_dict[name])
                elif isinstance(name, integer_types):
                    categorical_indices.add(name)
                else:
                    raise TypeError("Wrong type({}) or unknown name({}) in categorical_feature"
                                    .format(type(name).__name__, name))
1095
            if categorical_indices:
1096
1097
1098
1099
                for cat_alias in _ConfigAliases.get("categorical_feature"):
                    if cat_alias in params:
                        warnings.warn('{} in param dict is overridden.'.format(cat_alias))
                        params.pop(cat_alias, None)
1100
                params['categorical_column'] = sorted(categorical_indices)
1101

wxchan's avatar
wxchan committed
1102
        params_str = param_dict_to_str(params)
1103
        self.params = params
1104
        # process for reference dataset
wxchan's avatar
wxchan committed
1105
        ref_dataset = None
wxchan's avatar
wxchan committed
1106
        if isinstance(reference, Dataset):
1107
            ref_dataset = reference.construct().handle
wxchan's avatar
wxchan committed
1108
1109
        elif reference is not None:
            raise TypeError('Reference dataset should be None or dataset instance')
1110
        # start construct data
wxchan's avatar
wxchan committed
1111
        if isinstance(data, string_type):
wxchan's avatar
wxchan committed
1112
1113
1114
1115
1116
1117
1118
1119
            self.handle = ctypes.c_void_p()
            _safe_call(_LIB.LGBM_DatasetCreateFromFile(
                c_str(data),
                c_str(params_str),
                ref_dataset,
                ctypes.byref(self.handle)))
        elif isinstance(data, scipy.sparse.csr_matrix):
            self.__init_from_csr(data, params_str, ref_dataset)
Guolin Ke's avatar
Guolin Ke committed
1120
1121
        elif isinstance(data, scipy.sparse.csc_matrix):
            self.__init_from_csc(data, params_str, ref_dataset)
wxchan's avatar
wxchan committed
1122
1123
        elif isinstance(data, np.ndarray):
            self.__init_from_np2d(data, params_str, ref_dataset)
1124
1125
        elif isinstance(data, list) and len(data) > 0 and all(isinstance(x, np.ndarray) for x in data):
            self.__init_from_list_np2d(data, params_str, ref_dataset)
1126
1127
        elif isinstance(data, DataTable):
            self.__init_from_np2d(data.to_numpy(), params_str, ref_dataset)
wxchan's avatar
wxchan committed
1128
1129
1130
1131
        else:
            try:
                csr = scipy.sparse.csr_matrix(data)
                self.__init_from_csr(csr, params_str, ref_dataset)
1132
            except BaseException:
wxchan's avatar
wxchan committed
1133
                raise TypeError('Cannot initialize Dataset from {}'.format(type(data).__name__))
wxchan's avatar
wxchan committed
1134
1135
1136
        if label is not None:
            self.set_label(label)
        if self.get_label() is None:
1137
            raise ValueError("Label should not be None")
wxchan's avatar
wxchan committed
1138
1139
1140
1141
        if weight is not None:
            self.set_weight(weight)
        if group is not None:
            self.set_group(group)
1142
1143
1144
        if isinstance(predictor, _InnerPredictor):
            if self._predictor is None and init_score is not None:
                warnings.warn("The init_score will be overridden by the prediction of init_model.")
Guolin Ke's avatar
Guolin Ke committed
1145
            self._set_init_score_by_predictor(predictor, data)
1146
1147
        elif init_score is not None:
            self.set_init_score(init_score)
Guolin Ke's avatar
Guolin Ke committed
1148
1149
        elif predictor is not None:
            raise TypeError('Wrong predictor type {}'.format(type(predictor).__name__))
Guolin Ke's avatar
Guolin Ke committed
1150
        # set feature names
Nikita Titov's avatar
Nikita Titov committed
1151
        return self.set_feature_name(feature_name)
wxchan's avatar
wxchan committed
1152
1153

    def __init_from_np2d(self, mat, params_str, ref_dataset):
1154
        """Initialize data from a 2-D numpy matrix."""
wxchan's avatar
wxchan committed
1155
1156
1157
1158
1159
1160
        if len(mat.shape) != 2:
            raise ValueError('Input numpy.ndarray must be 2 dimensional')

        self.handle = ctypes.c_void_p()
        if mat.dtype == np.float32 or mat.dtype == np.float64:
            data = np.array(mat.reshape(mat.size), dtype=mat.dtype, copy=False)
1161
        else:  # change non-float data to float data, need to copy
wxchan's avatar
wxchan committed
1162
1163
            data = np.array(mat.reshape(mat.size), dtype=np.float32)

1164
        ptr_data, type_ptr_data, _ = c_float_array(data)
wxchan's avatar
wxchan committed
1165
1166
        _safe_call(_LIB.LGBM_DatasetCreateFromMat(
            ptr_data,
Guolin Ke's avatar
Guolin Ke committed
1167
1168
1169
1170
            ctypes.c_int(type_ptr_data),
            ctypes.c_int(mat.shape[0]),
            ctypes.c_int(mat.shape[1]),
            ctypes.c_int(C_API_IS_ROW_MAJOR),
wxchan's avatar
wxchan committed
1171
1172
1173
            c_str(params_str),
            ref_dataset,
            ctypes.byref(self.handle)))
Nikita Titov's avatar
Nikita Titov committed
1174
        return self
wxchan's avatar
wxchan committed
1175

1176
    def __init_from_list_np2d(self, mats, params_str, ref_dataset):
1177
        """Initialize data from a list of 2-D numpy matrices."""
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
        ncol = mats[0].shape[1]
        nrow = np.zeros((len(mats),), np.int32)
        if mats[0].dtype == np.float64:
            ptr_data = (ctypes.POINTER(ctypes.c_double) * len(mats))()
        else:
            ptr_data = (ctypes.POINTER(ctypes.c_float) * len(mats))()

        holders = []
        type_ptr_data = None

        for i, mat in enumerate(mats):
            if len(mat.shape) != 2:
                raise ValueError('Input numpy.ndarray must be 2 dimensional')

            if mat.shape[1] != ncol:
                raise ValueError('Input arrays must have same number of columns')

            nrow[i] = mat.shape[0]

            if mat.dtype == np.float32 or mat.dtype == np.float64:
                mats[i] = np.array(mat.reshape(mat.size), dtype=mat.dtype, copy=False)
1199
            else:  # change non-float data to float data, need to copy
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
                mats[i] = np.array(mat.reshape(mat.size), dtype=np.float32)

            chunk_ptr_data, chunk_type_ptr_data, holder = c_float_array(mats[i])
            if type_ptr_data is not None and chunk_type_ptr_data != type_ptr_data:
                raise ValueError('Input chunks must have same type')
            ptr_data[i] = chunk_ptr_data
            type_ptr_data = chunk_type_ptr_data
            holders.append(holder)

        self.handle = ctypes.c_void_p()
        _safe_call(_LIB.LGBM_DatasetCreateFromMats(
            ctypes.c_int(len(mats)),
            ctypes.cast(ptr_data, ctypes.POINTER(ctypes.POINTER(ctypes.c_double))),
            ctypes.c_int(type_ptr_data),
            nrow.ctypes.data_as(ctypes.POINTER(ctypes.c_int32)),
            ctypes.c_int(ncol),
            ctypes.c_int(C_API_IS_ROW_MAJOR),
            c_str(params_str),
            ref_dataset,
            ctypes.byref(self.handle)))
Nikita Titov's avatar
Nikita Titov committed
1220
        return self
1221

wxchan's avatar
wxchan committed
1222
    def __init_from_csr(self, csr, params_str, ref_dataset):
1223
        """Initialize data from a CSR matrix."""
wxchan's avatar
wxchan committed
1224
        if len(csr.indices) != len(csr.data):
1225
            raise ValueError('Length mismatch: {} vs {}'.format(len(csr.indices), len(csr.data)))
wxchan's avatar
wxchan committed
1226
1227
        self.handle = ctypes.c_void_p()

1228
1229
        ptr_indptr, type_ptr_indptr, __ = c_int_array(csr.indptr)
        ptr_data, type_ptr_data, _ = c_float_array(csr.data)
wxchan's avatar
wxchan committed
1230

1231
        assert csr.shape[1] <= MAX_INT32
1232
        csr_indices = csr.indices.astype(np.int32, copy=False)
1233

wxchan's avatar
wxchan committed
1234
1235
        _safe_call(_LIB.LGBM_DatasetCreateFromCSR(
            ptr_indptr,
Guolin Ke's avatar
Guolin Ke committed
1236
            ctypes.c_int(type_ptr_indptr),
1237
            csr_indices.ctypes.data_as(ctypes.POINTER(ctypes.c_int32)),
wxchan's avatar
wxchan committed
1238
            ptr_data,
Guolin Ke's avatar
Guolin Ke committed
1239
1240
1241
1242
            ctypes.c_int(type_ptr_data),
            ctypes.c_int64(len(csr.indptr)),
            ctypes.c_int64(len(csr.data)),
            ctypes.c_int64(csr.shape[1]),
wxchan's avatar
wxchan committed
1243
1244
1245
            c_str(params_str),
            ref_dataset,
            ctypes.byref(self.handle)))
Nikita Titov's avatar
Nikita Titov committed
1246
        return self
wxchan's avatar
wxchan committed
1247

Guolin Ke's avatar
Guolin Ke committed
1248
    def __init_from_csc(self, csc, params_str, ref_dataset):
1249
        """Initialize data from a CSC matrix."""
Guolin Ke's avatar
Guolin Ke committed
1250
1251
1252
1253
        if len(csc.indices) != len(csc.data):
            raise ValueError('Length mismatch: {} vs {}'.format(len(csc.indices), len(csc.data)))
        self.handle = ctypes.c_void_p()

1254
1255
        ptr_indptr, type_ptr_indptr, __ = c_int_array(csc.indptr)
        ptr_data, type_ptr_data, _ = c_float_array(csc.data)
Guolin Ke's avatar
Guolin Ke committed
1256

1257
        assert csc.shape[0] <= MAX_INT32
1258
        csc_indices = csc.indices.astype(np.int32, copy=False)
1259

Guolin Ke's avatar
Guolin Ke committed
1260
1261
        _safe_call(_LIB.LGBM_DatasetCreateFromCSC(
            ptr_indptr,
Guolin Ke's avatar
Guolin Ke committed
1262
            ctypes.c_int(type_ptr_indptr),
1263
            csc_indices.ctypes.data_as(ctypes.POINTER(ctypes.c_int32)),
Guolin Ke's avatar
Guolin Ke committed
1264
            ptr_data,
Guolin Ke's avatar
Guolin Ke committed
1265
1266
1267
1268
            ctypes.c_int(type_ptr_data),
            ctypes.c_int64(len(csc.indptr)),
            ctypes.c_int64(len(csc.data)),
            ctypes.c_int64(csc.shape[0]),
Guolin Ke's avatar
Guolin Ke committed
1269
1270
1271
            c_str(params_str),
            ref_dataset,
            ctypes.byref(self.handle)))
Nikita Titov's avatar
Nikita Titov committed
1272
        return self
Guolin Ke's avatar
Guolin Ke committed
1273

wxchan's avatar
wxchan committed
1274
    def construct(self):
1275
1276
1277
1278
1279
        """Lazy init.

        Returns
        -------
        self : Dataset
Nikita Titov's avatar
Nikita Titov committed
1280
            Constructed Dataset object.
1281
        """
1282
        if self.handle is None:
wxchan's avatar
wxchan committed
1283
            if self.reference is not None:
1284
1285
1286
1287
                reference_params = self.reference.get_params()
                if self.get_params() != reference_params:
                    warnings.warn('Overriding the parameters from Reference Dataset.')
                    self._update_params(reference_params)
wxchan's avatar
wxchan committed
1288
                if self.used_indices is None:
1289
                    # create valid
1290
                    self._lazy_init(self.data, label=self.label, reference=self.reference,
1291
1292
                                    weight=self.weight, group=self.group,
                                    init_score=self.init_score, predictor=self._predictor,
1293
                                    silent=self.silent, feature_name=self.feature_name, params=self.params)
wxchan's avatar
wxchan committed
1294
                else:
1295
                    # construct subset
wxchan's avatar
wxchan committed
1296
                    used_indices = list_to_1d_numpy(self.used_indices, np.int32, name='used_indices')
1297
                    assert used_indices.flags.c_contiguous
Guolin Ke's avatar
Guolin Ke committed
1298
                    if self.reference.group is not None:
1299
                        group_info = np.array(self.reference.group).astype(np.int32, copy=False)
1300
1301
                        _, self.group = np.unique(np.repeat(range_(len(group_info)), repeats=group_info)[self.used_indices],
                                                  return_counts=True)
1302
                    self.handle = ctypes.c_void_p()
wxchan's avatar
wxchan committed
1303
1304
                    params_str = param_dict_to_str(self.params)
                    _safe_call(_LIB.LGBM_DatasetGetSubset(
1305
                        self.reference.construct().handle,
wxchan's avatar
wxchan committed
1306
                        used_indices.ctypes.data_as(ctypes.POINTER(ctypes.c_int32)),
Guolin Ke's avatar
Guolin Ke committed
1307
                        ctypes.c_int(used_indices.shape[0]),
wxchan's avatar
wxchan committed
1308
1309
                        c_str(params_str),
                        ctypes.byref(self.handle)))
Guolin Ke's avatar
Guolin Ke committed
1310
1311
                    if not self.free_raw_data:
                        self.get_data()
Guolin Ke's avatar
Guolin Ke committed
1312
1313
                    if self.group is not None:
                        self.set_group(self.group)
wxchan's avatar
wxchan committed
1314
1315
                    if self.get_label() is None:
                        raise ValueError("Label should not be None.")
Guolin Ke's avatar
Guolin Ke committed
1316
1317
1318
                    if isinstance(self._predictor, _InnerPredictor) and self._predictor is not self.reference._predictor:
                        self.get_data()
                        self._set_init_score_by_predictor(self._predictor, self.data, used_indices)
wxchan's avatar
wxchan committed
1319
            else:
1320
                # create train
1321
                self._lazy_init(self.data, label=self.label,
1322
1323
1324
                                weight=self.weight, group=self.group,
                                init_score=self.init_score, predictor=self._predictor,
                                silent=self.silent, feature_name=self.feature_name,
1325
                                categorical_feature=self.categorical_feature, params=self.params)
wxchan's avatar
wxchan committed
1326
1327
1328
            if self.free_raw_data:
                self.data = None
        return self
wxchan's avatar
wxchan committed
1329

wxchan's avatar
wxchan committed
1330
    def create_valid(self, data, label=None, weight=None, group=None,
1331
                     init_score=None, silent=False, params=None):
1332
        """Create validation data align with current Dataset.
wxchan's avatar
wxchan committed
1333
1334
1335

        Parameters
        ----------
1336
        data : string, numpy array, pandas DataFrame, H2O DataTable's Frame, scipy.sparse or list of numpy arrays
wxchan's avatar
wxchan committed
1337
            Data source of Dataset.
1338
            If string, it represents the path to txt file.
1339
        label : list, numpy 1-D array, pandas Series / one-column DataFrame or None, optional (default=None)
1340
1341
            Label of the data.
        weight : list, numpy 1-D array, pandas Series or None, optional (default=None)
wxchan's avatar
wxchan committed
1342
            Weight for each instance.
1343
        group : list, numpy 1-D array, pandas Series or None, optional (default=None)
1344
            Group/query size for Dataset.
1345
        init_score : list, numpy 1-D array, pandas Series or None, optional (default=None)
1346
            Init score for Dataset.
1347
1348
        silent : bool, optional (default=False)
            Whether to print messages during construction.
Nikita Titov's avatar
Nikita Titov committed
1349
        params : dict or None, optional (default=None)
1350
            Other parameters for validation Dataset.
1351
1352
1353

        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
1354
1355
        valid : Dataset
            Validation Dataset with reference to self.
wxchan's avatar
wxchan committed
1356
        """
1357
        ret = Dataset(data, label=label, reference=self,
1358
1359
                      weight=weight, group=group, init_score=init_score,
                      silent=silent, params=params, free_raw_data=self.free_raw_data)
wxchan's avatar
wxchan committed
1360
        ret._predictor = self._predictor
1361
        ret.pandas_categorical = self.pandas_categorical
wxchan's avatar
wxchan committed
1362
        return ret
wxchan's avatar
wxchan committed
1363

wxchan's avatar
wxchan committed
1364
    def subset(self, used_indices, params=None):
1365
        """Get subset of current Dataset.
wxchan's avatar
wxchan committed
1366
1367
1368
1369

        Parameters
        ----------
        used_indices : list of int
1370
            Indices used to create the subset.
Nikita Titov's avatar
Nikita Titov committed
1371
        params : dict or None, optional (default=None)
1372
            These parameters will be passed to Dataset constructor.
1373
1374
1375
1376
1377

        Returns
        -------
        subset : Dataset
            Subset of the current Dataset.
wxchan's avatar
wxchan committed
1378
        """
wxchan's avatar
wxchan committed
1379
1380
        if params is None:
            params = self.params
wxchan's avatar
wxchan committed
1381
        ret = Dataset(None, reference=self, feature_name=self.feature_name,
1382
1383
                      categorical_feature=self.categorical_feature, params=params,
                      free_raw_data=self.free_raw_data)
wxchan's avatar
wxchan committed
1384
        ret._predictor = self._predictor
1385
        ret.pandas_categorical = self.pandas_categorical
1386
        ret.used_indices = sorted(used_indices)
wxchan's avatar
wxchan committed
1387
1388
1389
        return ret

    def save_binary(self, filename):
1390
        """Save Dataset to a binary file.
wxchan's avatar
wxchan committed
1391

1392
1393
1394
1395
1396
        .. note::

            Please note that `init_score` is not saved in binary file.
            If you need it, please set it again after loading Dataset.

wxchan's avatar
wxchan committed
1397
1398
1399
1400
        Parameters
        ----------
        filename : string
            Name of the output file.
Nikita Titov's avatar
Nikita Titov committed
1401
1402
1403
1404
1405

        Returns
        -------
        self : Dataset
            Returns self.
wxchan's avatar
wxchan committed
1406
1407
1408
1409
        """
        _safe_call(_LIB.LGBM_DatasetSaveBinary(
            self.construct().handle,
            c_str(filename)))
Nikita Titov's avatar
Nikita Titov committed
1410
        return self
wxchan's avatar
wxchan committed
1411
1412

    def _update_params(self, params):
1413
1414
        if not params:
            return self
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
        params = copy.deepcopy(params)

        def update():
            if not self.params:
                self.params = params
            else:
                self.params_back_up = copy.deepcopy(self.params)
                self.params.update(params)

        if self.handle is None:
            update()
        elif params is not None:
            ret = _LIB.LGBM_DatasetUpdateParamChecking(
                c_str(param_dict_to_str(self.params)),
                c_str(param_dict_to_str(params)))
            if ret != 0:
                # could be updated if data is not freed
                if self.data is not None:
                    update()
                    self._free_handle()
                else:
                    raise LightGBMError(decode_string(_LIB.LGBM_GetLastError()))
Nikita Titov's avatar
Nikita Titov committed
1437
        return self
wxchan's avatar
wxchan committed
1438

1439
    def _reverse_update_params(self):
1440
1441
1442
        if self.handle is None:
            self.params = copy.deepcopy(self.params_back_up)
            self.params_back_up = None
Nikita Titov's avatar
Nikita Titov committed
1443
        return self
1444

wxchan's avatar
wxchan committed
1445
    def set_field(self, field_name, data):
wxchan's avatar
wxchan committed
1446
        """Set property into the Dataset.
wxchan's avatar
wxchan committed
1447
1448
1449

        Parameters
        ----------
Nikita Titov's avatar
Nikita Titov committed
1450
        field_name : string
1451
            The field name of the information.
1452
        data : list, numpy 1-D array, pandas Series or None
1453
            The array of data to be set.
Nikita Titov's avatar
Nikita Titov committed
1454
1455
1456
1457
1458

        Returns
        -------
        self : Dataset
            Dataset with set property.
wxchan's avatar
wxchan committed
1459
        """
1460
1461
        if self.handle is None:
            raise Exception("Cannot set %s before construct dataset" % field_name)
wxchan's avatar
wxchan committed
1462
        if data is None:
1463
            # set to None
wxchan's avatar
wxchan committed
1464
1465
1466
1467
            _safe_call(_LIB.LGBM_DatasetSetField(
                self.handle,
                c_str(field_name),
                None,
Guolin Ke's avatar
Guolin Ke committed
1468
1469
                ctypes.c_int(0),
                ctypes.c_int(FIELD_TYPE_MAPPER[field_name])))
Nikita Titov's avatar
Nikita Titov committed
1470
            return self
Guolin Ke's avatar
Guolin Ke committed
1471
1472
1473
1474
1475
        dtype = np.float32
        if field_name == 'group':
            dtype = np.int32
        elif field_name == 'init_score':
            dtype = np.float64
1476
        data = list_to_1d_numpy(data, dtype, name=field_name)
1477
1478
        if data.dtype == np.float32 or data.dtype == np.float64:
            ptr_data, type_data, _ = c_float_array(data)
wxchan's avatar
wxchan committed
1479
        elif data.dtype == np.int32:
1480
            ptr_data, type_data, _ = c_int_array(data)
wxchan's avatar
wxchan committed
1481
        else:
Nikita Titov's avatar
Nikita Titov committed
1482
            raise TypeError("Expected np.float32/64 or np.int32, met type({})".format(data.dtype))
wxchan's avatar
wxchan committed
1483
        if type_data != FIELD_TYPE_MAPPER[field_name]:
1484
            raise TypeError("Input type error for set_field")
wxchan's avatar
wxchan committed
1485
1486
1487
1488
        _safe_call(_LIB.LGBM_DatasetSetField(
            self.handle,
            c_str(field_name),
            ptr_data,
Guolin Ke's avatar
Guolin Ke committed
1489
1490
            ctypes.c_int(len(data)),
            ctypes.c_int(type_data)))
1491
        self.version += 1
Nikita Titov's avatar
Nikita Titov committed
1492
        return self
wxchan's avatar
wxchan committed
1493

wxchan's avatar
wxchan committed
1494
1495
    def get_field(self, field_name):
        """Get property from the Dataset.
wxchan's avatar
wxchan committed
1496
1497
1498

        Parameters
        ----------
Nikita Titov's avatar
Nikita Titov committed
1499
        field_name : string
1500
            The field name of the information.
wxchan's avatar
wxchan committed
1501
1502
1503

        Returns
        -------
1504
1505
        info : numpy array
            A numpy array with information from the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1506
        """
1507
        if self.handle is None:
1508
            raise Exception("Cannot get %s before construct Dataset" % field_name)
Guolin Ke's avatar
Guolin Ke committed
1509
1510
        tmp_out_len = ctypes.c_int()
        out_type = ctypes.c_int()
wxchan's avatar
wxchan committed
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
        ret = ctypes.POINTER(ctypes.c_void_p)()
        _safe_call(_LIB.LGBM_DatasetGetField(
            self.handle,
            c_str(field_name),
            ctypes.byref(tmp_out_len),
            ctypes.byref(ret),
            ctypes.byref(out_type)))
        if out_type.value != FIELD_TYPE_MAPPER[field_name]:
            raise TypeError("Return type error for get_field")
        if tmp_out_len.value == 0:
            return None
        if out_type.value == C_API_DTYPE_INT32:
            return cint32_array_to_numpy(ctypes.cast(ret, ctypes.POINTER(ctypes.c_int32)), tmp_out_len.value)
        elif out_type.value == C_API_DTYPE_FLOAT32:
            return cfloat32_array_to_numpy(ctypes.cast(ret, ctypes.POINTER(ctypes.c_float)), tmp_out_len.value)
Guolin Ke's avatar
Guolin Ke committed
1526
1527
        elif out_type.value == C_API_DTYPE_FLOAT64:
            return cfloat64_array_to_numpy(ctypes.cast(ret, ctypes.POINTER(ctypes.c_double)), tmp_out_len.value)
1528
        else:
wxchan's avatar
wxchan committed
1529
            raise TypeError("Unknown type")
Guolin Ke's avatar
Guolin Ke committed
1530

1531
    def set_categorical_feature(self, categorical_feature):
1532
        """Set categorical features.
1533
1534
1535

        Parameters
        ----------
1536
1537
        categorical_feature : list of int or strings
            Names or indices of categorical features.
Nikita Titov's avatar
Nikita Titov committed
1538
1539
1540
1541
1542

        Returns
        -------
        self : Dataset
            Dataset with set categorical features.
1543
1544
        """
        if self.categorical_feature == categorical_feature:
Nikita Titov's avatar
Nikita Titov committed
1545
            return self
1546
        if self.data is not None:
1547
1548
            if self.categorical_feature is None:
                self.categorical_feature = categorical_feature
Nikita Titov's avatar
Nikita Titov committed
1549
                return self._free_handle()
1550
1551
            elif categorical_feature == 'auto':
                warnings.warn('Using categorical_feature in Dataset.')
Nikita Titov's avatar
Nikita Titov committed
1552
                return self
1553
            else:
1554
1555
                warnings.warn('categorical_feature in Dataset is overridden.\n'
                              'New categorical_feature is {}'.format(sorted(list(categorical_feature))))
1556
                self.categorical_feature = categorical_feature
Nikita Titov's avatar
Nikita Titov committed
1557
                return self._free_handle()
1558
        else:
1559
1560
            raise LightGBMError("Cannot set categorical feature after freed raw data, "
                                "set free_raw_data=False when construct Dataset to avoid this.")
1561

Guolin Ke's avatar
Guolin Ke committed
1562
    def _set_predictor(self, predictor):
1563
1564
1565
1566
        """Set predictor for continued training.

        It is not recommended for user to call this function.
        Please use init_model argument in engine.train() or engine.cv() instead.
Guolin Ke's avatar
Guolin Ke committed
1567
        """
1568
        if predictor is self._predictor and (predictor is None or predictor.current_iteration() == self._predictor.current_iteration()):
Nikita Titov's avatar
Nikita Titov committed
1569
            return self
1570
        if self.handle is None:
Guolin Ke's avatar
Guolin Ke committed
1571
            self._predictor = predictor
1572
1573
1574
1575
1576
1577
        elif self.data is not None:
            self._predictor = predictor
            self._set_init_score_by_predictor(self._predictor, self.data)
        elif self.used_indices is not None and self.reference is not None and self.reference.data is not None:
            self._predictor = predictor
            self._set_init_score_by_predictor(self._predictor, self.reference.data, self.used_indices)
Guolin Ke's avatar
Guolin Ke committed
1578
        else:
1579
1580
            raise LightGBMError("Cannot set predictor after freed raw data, "
                                "set free_raw_data=False when construct Dataset to avoid this.")
1581
        return self
Guolin Ke's avatar
Guolin Ke committed
1582
1583

    def set_reference(self, reference):
1584
        """Set reference Dataset.
Guolin Ke's avatar
Guolin Ke committed
1585
1586
1587
1588

        Parameters
        ----------
        reference : Dataset
1589
            Reference that is used as a template to construct the current Dataset.
Nikita Titov's avatar
Nikita Titov committed
1590
1591
1592
1593
1594

        Returns
        -------
        self : Dataset
            Dataset with set reference.
Guolin Ke's avatar
Guolin Ke committed
1595
        """
1596
1597
1598
        self.set_categorical_feature(reference.categorical_feature) \
            .set_feature_name(reference.feature_name) \
            ._set_predictor(reference._predictor)
1599
1600
        # we're done if self and reference share a common upstrem reference
        if self.get_ref_chain().intersection(reference.get_ref_chain()):
Nikita Titov's avatar
Nikita Titov committed
1601
            return self
Guolin Ke's avatar
Guolin Ke committed
1602
1603
        if self.data is not None:
            self.reference = reference
Nikita Titov's avatar
Nikita Titov committed
1604
            return self._free_handle()
Guolin Ke's avatar
Guolin Ke committed
1605
        else:
1606
1607
            raise LightGBMError("Cannot set reference after freed raw data, "
                                "set free_raw_data=False when construct Dataset to avoid this.")
Guolin Ke's avatar
Guolin Ke committed
1608
1609

    def set_feature_name(self, feature_name):
1610
        """Set feature name.
Guolin Ke's avatar
Guolin Ke committed
1611
1612
1613

        Parameters
        ----------
1614
1615
        feature_name : list of strings
            Feature names.
Nikita Titov's avatar
Nikita Titov committed
1616
1617
1618
1619
1620

        Returns
        -------
        self : Dataset
            Dataset with set feature name.
Guolin Ke's avatar
Guolin Ke committed
1621
        """
1622
1623
        if feature_name != 'auto':
            self.feature_name = feature_name
1624
        if self.handle is not None and feature_name is not None and feature_name != 'auto':
wxchan's avatar
wxchan committed
1625
            if len(feature_name) != self.num_feature():
1626
1627
                raise ValueError("Length of feature_name({}) and num_feature({}) don't match"
                                 .format(len(feature_name), self.num_feature()))
1628
            c_feature_name = [c_str(name) for name in feature_name]
wxchan's avatar
wxchan committed
1629
1630
1631
            _safe_call(_LIB.LGBM_DatasetSetFeatureNames(
                self.handle,
                c_array(ctypes.c_char_p, c_feature_name),
Guolin Ke's avatar
Guolin Ke committed
1632
                ctypes.c_int(len(feature_name))))
Nikita Titov's avatar
Nikita Titov committed
1633
        return self
Guolin Ke's avatar
Guolin Ke committed
1634
1635

    def set_label(self, label):
1636
        """Set label of Dataset.
Guolin Ke's avatar
Guolin Ke committed
1637
1638
1639

        Parameters
        ----------
1640
        label : list, numpy 1-D array, pandas Series / one-column DataFrame or None
1641
            The label information to be set into Dataset.
Nikita Titov's avatar
Nikita Titov committed
1642
1643
1644
1645
1646

        Returns
        -------
        self : Dataset
            Dataset with set label.
Guolin Ke's avatar
Guolin Ke committed
1647
1648
        """
        self.label = label
1649
        if self.handle is not None:
1650
            label = list_to_1d_numpy(_label_from_pandas(label), name='label')
wxchan's avatar
wxchan committed
1651
            self.set_field('label', label)
1652
            self.label = self.get_field('label')  # original values can be modified at cpp side
Nikita Titov's avatar
Nikita Titov committed
1653
        return self
Guolin Ke's avatar
Guolin Ke committed
1654
1655

    def set_weight(self, weight):
1656
        """Set weight of each instance.
Guolin Ke's avatar
Guolin Ke committed
1657
1658
1659

        Parameters
        ----------
1660
        weight : list, numpy 1-D array, pandas Series or None
1661
            Weight to be set for each data point.
Nikita Titov's avatar
Nikita Titov committed
1662
1663
1664
1665
1666

        Returns
        -------
        self : Dataset
            Dataset with set weight.
Guolin Ke's avatar
Guolin Ke committed
1667
        """
1668
1669
        if weight is not None and np.all(weight == 1):
            weight = None
Guolin Ke's avatar
Guolin Ke committed
1670
        self.weight = weight
1671
        if self.handle is not None and weight is not None:
wxchan's avatar
wxchan committed
1672
1673
            weight = list_to_1d_numpy(weight, name='weight')
            self.set_field('weight', weight)
1674
            self.weight = self.get_field('weight')  # original values can be modified at cpp side
Nikita Titov's avatar
Nikita Titov committed
1675
        return self
Guolin Ke's avatar
Guolin Ke committed
1676
1677

    def set_init_score(self, init_score):
1678
        """Set init score of Booster to start from.
Guolin Ke's avatar
Guolin Ke committed
1679
1680
1681

        Parameters
        ----------
1682
        init_score : list, numpy 1-D array, pandas Series or None
1683
            Init score for Booster.
Nikita Titov's avatar
Nikita Titov committed
1684
1685
1686
1687
1688

        Returns
        -------
        self : Dataset
            Dataset with set init score.
Guolin Ke's avatar
Guolin Ke committed
1689
1690
        """
        self.init_score = init_score
1691
        if self.handle is not None and init_score is not None:
Guolin Ke's avatar
Guolin Ke committed
1692
            init_score = list_to_1d_numpy(init_score, np.float64, name='init_score')
wxchan's avatar
wxchan committed
1693
            self.set_field('init_score', init_score)
1694
            self.init_score = self.get_field('init_score')  # original values can be modified at cpp side
Nikita Titov's avatar
Nikita Titov committed
1695
        return self
Guolin Ke's avatar
Guolin Ke committed
1696
1697

    def set_group(self, group):
1698
        """Set group size of Dataset (used for ranking).
Guolin Ke's avatar
Guolin Ke committed
1699
1700
1701

        Parameters
        ----------
1702
        group : list, numpy 1-D array, pandas Series or None
1703
            Group size of each group.
Nikita Titov's avatar
Nikita Titov committed
1704
1705
1706
1707
1708

        Returns
        -------
        self : Dataset
            Dataset with set group.
Guolin Ke's avatar
Guolin Ke committed
1709
1710
        """
        self.group = group
1711
        if self.handle is not None and group is not None:
wxchan's avatar
wxchan committed
1712
1713
            group = list_to_1d_numpy(group, np.int32, name='group')
            self.set_field('group', group)
Nikita Titov's avatar
Nikita Titov committed
1714
        return self
Guolin Ke's avatar
Guolin Ke committed
1715

1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
    def get_feature_name(self):
        """Get the names of columns (features) in the Dataset.

        Returns
        -------
        feature_names : list
            The names of columns (features) in the Dataset.
        """
        if self.handle is None:
            raise LightGBMError("Cannot get feature_name before construct dataset")
        num_feature = self.num_feature()
        tmp_out_len = ctypes.c_int(0)
        reserved_string_buffer_size = 255
        required_string_buffer_size = ctypes.c_size_t(0)
        string_buffers = [ctypes.create_string_buffer(reserved_string_buffer_size) for i in range_(num_feature)]
        ptr_string_buffers = (ctypes.c_char_p * num_feature)(*map(ctypes.addressof, string_buffers))
        _safe_call(_LIB.LGBM_DatasetGetFeatureNames(
            self.handle,
            num_feature,
            ctypes.byref(tmp_out_len),
            reserved_string_buffer_size,
            ctypes.byref(required_string_buffer_size),
            ptr_string_buffers))
        if num_feature != tmp_out_len.value:
            raise ValueError("Length of feature names doesn't equal with num_feature")
        if reserved_string_buffer_size < required_string_buffer_size.value:
            raise BufferError(
                "Allocated feature name buffer size ({}) was inferior to the needed size ({})."
                .format(reserved_string_buffer_size, required_string_buffer_size.value)
            )
        return [string_buffers[i].value.decode('utf-8') for i in range_(num_feature)]

Guolin Ke's avatar
Guolin Ke committed
1748
    def get_label(self):
1749
        """Get the label of the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1750
1751
1752

        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
1753
        label : numpy array or None
1754
            The label information from the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1755
        """
1756
        if self.label is None:
wxchan's avatar
wxchan committed
1757
            self.label = self.get_field('label')
Guolin Ke's avatar
Guolin Ke committed
1758
1759
1760
        return self.label

    def get_weight(self):
1761
        """Get the weight of the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1762
1763
1764

        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
1765
        weight : numpy array or None
1766
            Weight for each data point from the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1767
        """
1768
        if self.weight is None:
wxchan's avatar
wxchan committed
1769
            self.weight = self.get_field('weight')
Guolin Ke's avatar
Guolin Ke committed
1770
1771
1772
        return self.weight

    def get_init_score(self):
1773
        """Get the initial score of the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1774
1775
1776

        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
1777
        init_score : numpy array or None
1778
            Init score of Booster.
Guolin Ke's avatar
Guolin Ke committed
1779
        """
1780
        if self.init_score is None:
wxchan's avatar
wxchan committed
1781
            self.init_score = self.get_field('init_score')
Guolin Ke's avatar
Guolin Ke committed
1782
1783
        return self.init_score

1784
1785
1786
1787
1788
    def get_data(self):
        """Get the raw data of the Dataset.

        Returns
        -------
1789
        data : string, numpy array, pandas DataFrame, H2O DataTable's Frame, scipy.sparse, list of numpy arrays or None
1790
1791
1792
1793
            Raw data used in the Dataset construction.
        """
        if self.handle is None:
            raise Exception("Cannot get data before construct Dataset")
Guolin Ke's avatar
Guolin Ke committed
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
        if self.need_slice and self.used_indices is not None and self.reference is not None:
            self.data = self.reference.data
            if self.data is not None:
                if isinstance(self.data, np.ndarray) or scipy.sparse.issparse(self.data):
                    self.data = self.data[self.used_indices, :]
                elif isinstance(self.data, DataFrame):
                    self.data = self.data.iloc[self.used_indices].copy()
                elif isinstance(self.data, DataTable):
                    self.data = self.data[self.used_indices, :]
                else:
                    warnings.warn("Cannot subset {} type of raw data.\n"
                                  "Returning original raw data".format(type(self.data).__name__))
1806
            self.need_slice = False
Guolin Ke's avatar
Guolin Ke committed
1807
1808
1809
        if self.data is None:
            raise LightGBMError("Cannot call `get_data` after freed raw data, "
                                "set free_raw_data=False when construct Dataset to avoid this.")
1810
1811
        return self.data

Guolin Ke's avatar
Guolin Ke committed
1812
    def get_group(self):
1813
        """Get the group of the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1814
1815
1816

        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
1817
        group : numpy array or None
1818
            Group size of each group.
Guolin Ke's avatar
Guolin Ke committed
1819
        """
1820
        if self.group is None:
wxchan's avatar
wxchan committed
1821
            self.group = self.get_field('group')
Guolin Ke's avatar
Guolin Ke committed
1822
1823
            if self.group is not None:
                # group data from LightGBM is boundaries data, need to convert to group size
Nikita Titov's avatar
Nikita Titov committed
1824
                self.group = np.diff(self.group)
Guolin Ke's avatar
Guolin Ke committed
1825
1826
1827
        return self.group

    def num_data(self):
1828
        """Get the number of rows in the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1829
1830
1831

        Returns
        -------
1832
1833
        number_of_rows : int
            The number of rows in the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1834
        """
1835
        if self.handle is not None:
Guolin Ke's avatar
Guolin Ke committed
1836
            ret = ctypes.c_int()
wxchan's avatar
wxchan committed
1837
1838
1839
            _safe_call(_LIB.LGBM_DatasetGetNumData(self.handle,
                                                   ctypes.byref(ret)))
            return ret.value
Guolin Ke's avatar
Guolin Ke committed
1840
        else:
1841
            raise LightGBMError("Cannot get num_data before construct dataset")
Guolin Ke's avatar
Guolin Ke committed
1842
1843

    def num_feature(self):
1844
        """Get the number of columns (features) in the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1845
1846
1847

        Returns
        -------
1848
1849
        number_of_columns : int
            The number of columns (features) in the Dataset.
Guolin Ke's avatar
Guolin Ke committed
1850
        """
1851
        if self.handle is not None:
Guolin Ke's avatar
Guolin Ke committed
1852
            ret = ctypes.c_int()
wxchan's avatar
wxchan committed
1853
1854
1855
            _safe_call(_LIB.LGBM_DatasetGetNumFeature(self.handle,
                                                      ctypes.byref(ret)))
            return ret.value
Guolin Ke's avatar
Guolin Ke committed
1856
        else:
1857
            raise LightGBMError("Cannot get num_feature before construct dataset")
Guolin Ke's avatar
Guolin Ke committed
1858

1859
    def get_ref_chain(self, ref_limit=100):
1860
1861
1862
1863
1864
        """Get a chain of Dataset objects.

        Starts with r, then goes to r.reference (if exists),
        then to r.reference.reference, etc.
        until we hit ``ref_limit`` or a reference loop.
1865
1866
1867
1868
1869

        Parameters
        ----------
        ref_limit : int, optional (default=100)
            The limit number of references.
1870
1871
1872

        Returns
        -------
1873
1874
1875
        ref_chain : set of Dataset
            Chain of references of the Datasets.
        """
1876
        head = self
1877
        ref_chain = set()
1878
1879
        while len(ref_chain) < ref_limit:
            if isinstance(head, Dataset):
1880
                ref_chain.add(head)
1881
1882
1883
1884
1885
1886
                if (head.reference is not None) and (head.reference not in ref_chain):
                    head = head.reference
                else:
                    break
            else:
                break
Nikita Titov's avatar
Nikita Titov committed
1887
        return ref_chain
1888

1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
    def add_features_from(self, other):
        """Add features from other Dataset to the current Dataset.

        Both Datasets must be constructed before calling this method.

        Parameters
        ----------
        other : Dataset
            The Dataset to take features from.

        Returns
        -------
        self : Dataset
            Dataset with the new features added.
        """
        if self.handle is None or other.handle is None:
            raise ValueError('Both source and target Datasets must be constructed before adding features')
        _safe_call(_LIB.LGBM_DatasetAddFeaturesFrom(self.handle, other.handle))
        return self

1909
    def _dump_text(self, filename):
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
        """Save Dataset to a text file.

        This format cannot be loaded back in by LightGBM, but is useful for debugging purposes.

        Parameters
        ----------
        filename : string
            Name of the output file.

        Returns
        -------
        self : Dataset
            Returns self.
        """
        _safe_call(_LIB.LGBM_DatasetDumpText(
            self.construct().handle,
            c_str(filename)))
        return self

wxchan's avatar
wxchan committed
1929

wxchan's avatar
wxchan committed
1930
class Booster(object):
1931
    """Booster in LightGBM."""
1932

1933
    def __init__(self, params=None, train_set=None, model_file=None, model_str=None, silent=False):
1934
        """Initialize the Booster.
wxchan's avatar
wxchan committed
1935
1936
1937

        Parameters
        ----------
Nikita Titov's avatar
Nikita Titov committed
1938
        params : dict or None, optional (default=None)
1939
1940
1941
1942
            Parameters for Booster.
        train_set : Dataset or None, optional (default=None)
            Training dataset.
        model_file : string or None, optional (default=None)
wxchan's avatar
wxchan committed
1943
            Path to the model file.
1944
1945
        model_str : string or None, optional (default=None)
            Model will be loaded from this string.
1946
1947
        silent : bool, optional (default=False)
            Whether to print messages during construction.
wxchan's avatar
wxchan committed
1948
        """
1949
        self.handle = None
1950
        self.network = False
wxchan's avatar
wxchan committed
1951
        self.__need_reload_eval_info = True
1952
        self._train_data_name = "training"
wxchan's avatar
wxchan committed
1953
        self.__attr = {}
1954
        self.__set_objective_to_none = False
wxchan's avatar
wxchan committed
1955
        self.best_iteration = -1
wxchan's avatar
wxchan committed
1956
        self.best_score = {}
1957
        params = {} if params is None else copy.deepcopy(params)
1958
        # user can set verbose with params, it has higher priority
1959
        if not any(verbose_alias in params for verbose_alias in _ConfigAliases.get("verbosity")) and silent:
1960
            params["verbose"] = -1
wxchan's avatar
wxchan committed
1961
        if train_set is not None:
1962
            # Training task
wxchan's avatar
wxchan committed
1963
            if not isinstance(train_set, Dataset):
1964
1965
                raise TypeError('Training data should be Dataset instance, met {}'
                                .format(type(train_set).__name__))
1966
            # set network if necessary
1967
            for alias in _ConfigAliases.get("machines"):
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
                if alias in params:
                    machines = params[alias]
                    if isinstance(machines, string_type):
                        num_machines = len(machines.split(','))
                    elif isinstance(machines, (list, set)):
                        num_machines = len(machines)
                        machines = ','.join(machines)
                    else:
                        raise ValueError("Invalid machines in params.")
                    self.set_network(machines,
                                     local_listen_port=params.get("local_listen_port", 12400),
                                     listen_time_out=params.get("listen_time_out", 120),
1980
                                     num_machines=params.setdefault("num_machines", num_machines))
1981
                    break
1982
            # construct booster object
1983
1984
1985
1986
            train_set.construct()
            # copy the parameters from train_set
            params.update(train_set.get_params())
            params_str = param_dict_to_str(params)
1987
            self.handle = ctypes.c_void_p()
wxchan's avatar
wxchan committed
1988
            _safe_call(_LIB.LGBM_BoosterCreate(
1989
                train_set.handle,
wxchan's avatar
wxchan committed
1990
1991
                c_str(params_str),
                ctypes.byref(self.handle)))
1992
            # save reference to data
wxchan's avatar
wxchan committed
1993
1994
1995
1996
            self.train_set = train_set
            self.valid_sets = []
            self.name_valid_sets = []
            self.__num_dataset = 1
Guolin Ke's avatar
Guolin Ke committed
1997
1998
            self.__init_predictor = train_set._predictor
            if self.__init_predictor is not None:
wxchan's avatar
wxchan committed
1999
2000
                _safe_call(_LIB.LGBM_BoosterMerge(
                    self.handle,
Guolin Ke's avatar
Guolin Ke committed
2001
                    self.__init_predictor.handle))
Guolin Ke's avatar
Guolin Ke committed
2002
            out_num_class = ctypes.c_int(0)
wxchan's avatar
wxchan committed
2003
2004
2005
2006
            _safe_call(_LIB.LGBM_BoosterGetNumClasses(
                self.handle,
                ctypes.byref(out_num_class)))
            self.__num_class = out_num_class.value
2007
            # buffer for inner predict
wxchan's avatar
wxchan committed
2008
2009
2010
            self.__inner_predict_buffer = [None]
            self.__is_predicted_cur_iter = [False]
            self.__get_eval_info()
2011
            self.pandas_categorical = train_set.pandas_categorical
2012
            self.train_set_version = train_set.version
wxchan's avatar
wxchan committed
2013
        elif model_file is not None:
2014
            # Prediction task
Guolin Ke's avatar
Guolin Ke committed
2015
            out_num_iterations = ctypes.c_int(0)
2016
            self.handle = ctypes.c_void_p()
wxchan's avatar
wxchan committed
2017
2018
2019
2020
            _safe_call(_LIB.LGBM_BoosterCreateFromModelfile(
                c_str(model_file),
                ctypes.byref(out_num_iterations),
                ctypes.byref(self.handle)))
Guolin Ke's avatar
Guolin Ke committed
2021
            out_num_class = ctypes.c_int(0)
wxchan's avatar
wxchan committed
2022
2023
2024
2025
            _safe_call(_LIB.LGBM_BoosterGetNumClasses(
                self.handle,
                ctypes.byref(out_num_class)))
            self.__num_class = out_num_class.value
2026
            self.pandas_categorical = _load_pandas_categorical(file_name=model_file)
2027
2028
        elif model_str is not None:
            self.model_from_string(model_str, not silent)
wxchan's avatar
wxchan committed
2029
        else:
2030
2031
            raise TypeError('Need at least one training dataset or model file or model string '
                            'to create Booster instance')
2032
        self.params = params
wxchan's avatar
wxchan committed
2033
2034

    def __del__(self):
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
        try:
            if self.network:
                self.free_network()
        except AttributeError:
            pass
        try:
            if self.handle is not None:
                _safe_call(_LIB.LGBM_BoosterFree(self.handle))
        except AttributeError:
            pass
wxchan's avatar
wxchan committed
2045

wxchan's avatar
wxchan committed
2046
2047
2048
2049
    def __copy__(self):
        return self.__deepcopy__(None)

    def __deepcopy__(self, _):
2050
        model_str = self.model_to_string(num_iteration=-1)
2051
        booster = Booster(model_str=model_str)
2052
        return booster
wxchan's avatar
wxchan committed
2053
2054
2055
2056
2057
2058
2059

    def __getstate__(self):
        this = self.__dict__.copy()
        handle = this['handle']
        this.pop('train_set', None)
        this.pop('valid_sets', None)
        if handle is not None:
2060
            this["handle"] = self.model_to_string(num_iteration=-1)
wxchan's avatar
wxchan committed
2061
2062
2063
        return this

    def __setstate__(self, state):
2064
2065
        model_str = state.get('handle', None)
        if model_str is not None:
wxchan's avatar
wxchan committed
2066
            handle = ctypes.c_void_p()
Guolin Ke's avatar
Guolin Ke committed
2067
            out_num_iterations = ctypes.c_int(0)
2068
2069
2070
2071
            _safe_call(_LIB.LGBM_BoosterLoadModelFromString(
                c_str(model_str),
                ctypes.byref(out_num_iterations),
                ctypes.byref(handle)))
wxchan's avatar
wxchan committed
2072
2073
2074
            state['handle'] = handle
        self.__dict__.update(state)

wxchan's avatar
wxchan committed
2075
    def free_dataset(self):
Nikita Titov's avatar
Nikita Titov committed
2076
2077
2078
2079
2080
2081
2082
        """Free Booster's Datasets.

        Returns
        -------
        self : Booster
            Booster without Datasets.
        """
wxchan's avatar
wxchan committed
2083
2084
        self.__dict__.pop('train_set', None)
        self.__dict__.pop('valid_sets', None)
2085
        self.__num_dataset = 0
Nikita Titov's avatar
Nikita Titov committed
2086
        return self
wxchan's avatar
wxchan committed
2087

2088
2089
2090
    def _free_buffer(self):
        self.__inner_predict_buffer = []
        self.__is_predicted_cur_iter = []
Nikita Titov's avatar
Nikita Titov committed
2091
        return self
2092

2093
2094
2095
2096
2097
2098
    def set_network(self, machines, local_listen_port=12400,
                    listen_time_out=120, num_machines=1):
        """Set the network configuration.

        Parameters
        ----------
Nikita Titov's avatar
Nikita Titov committed
2099
        machines : list, set or string
2100
            Names of machines.
Nikita Titov's avatar
Nikita Titov committed
2101
        local_listen_port : int, optional (default=12400)
2102
            TCP listen port for local machines.
Nikita Titov's avatar
Nikita Titov committed
2103
        listen_time_out : int, optional (default=120)
2104
            Socket time-out in minutes.
Nikita Titov's avatar
Nikita Titov committed
2105
        num_machines : int, optional (default=1)
2106
            The number of machines for parallel learning application.
Nikita Titov's avatar
Nikita Titov committed
2107
2108
2109
2110
2111

        Returns
        -------
        self : Booster
            Booster with set network.
2112
2113
2114
2115
2116
2117
        """
        _safe_call(_LIB.LGBM_NetworkInit(c_str(machines),
                                         ctypes.c_int(local_listen_port),
                                         ctypes.c_int(listen_time_out),
                                         ctypes.c_int(num_machines)))
        self.network = True
Nikita Titov's avatar
Nikita Titov committed
2118
        return self
2119
2120

    def free_network(self):
Nikita Titov's avatar
Nikita Titov committed
2121
2122
2123
2124
2125
2126
2127
        """Free Booster's network.

        Returns
        -------
        self : Booster
            Booster with freed network.
        """
2128
2129
        _safe_call(_LIB.LGBM_NetworkFree())
        self.network = False
Nikita Titov's avatar
Nikita Titov committed
2130
        return self
2131

2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
    def trees_to_dataframe(self):
        """Parse the fitted model and return in an easy-to-read pandas DataFrame.

        Returns
        -------
        result : pandas DataFrame
            Returns a pandas DataFrame of the parsed model.
        """
        if not PANDAS_INSTALLED:
            raise LightGBMError('This method cannot be run without pandas installed')

        if self.num_trees() == 0:
            raise LightGBMError('There are no trees in this Booster and thus nothing to parse')

        def _is_split_node(tree):
            return 'split_index' in tree.keys()

        def create_node_record(tree, node_depth=1, tree_index=None,
                               feature_names=None, parent_node=None):

            def _get_node_index(tree, tree_index):
                tree_num = str(tree_index) + '-' if tree_index is not None else ''
                is_split = _is_split_node(tree)
                node_type = 'S' if is_split else 'L'
                # if a single node tree it won't have `leaf_index` so return 0
                node_num = str(tree.get('split_index' if is_split else 'leaf_index', 0))
                return tree_num + node_type + node_num

            def _get_split_feature(tree, feature_names):
                if _is_split_node(tree):
                    if feature_names is not None:
                        feature_name = feature_names[tree['split_feature']]
                    else:
                        feature_name = tree['split_feature']
                else:
                    feature_name = None
                return feature_name

            def _is_single_node_tree(tree):
2171
                return set(tree.keys()) == {'leaf_value'}
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246

            # Create the node record, and populate universal data members
            node = OrderedDict()
            node['tree_index'] = tree_index
            node['node_depth'] = node_depth
            node['node_index'] = _get_node_index(tree, tree_index)
            node['left_child'] = None
            node['right_child'] = None
            node['parent_index'] = parent_node
            node['split_feature'] = _get_split_feature(tree, feature_names)
            node['split_gain'] = None
            node['threshold'] = None
            node['decision_type'] = None
            node['missing_direction'] = None
            node['missing_type'] = None
            node['value'] = None
            node['weight'] = None
            node['count'] = None

            # Update values to reflect node type (leaf or split)
            if _is_split_node(tree):
                node['left_child'] = _get_node_index(tree['left_child'], tree_index)
                node['right_child'] = _get_node_index(tree['right_child'], tree_index)
                node['split_gain'] = tree['split_gain']
                node['threshold'] = tree['threshold']
                node['decision_type'] = tree['decision_type']
                node['missing_direction'] = 'left' if tree['default_left'] else 'right'
                node['missing_type'] = tree['missing_type']
                node['value'] = tree['internal_value']
                node['weight'] = tree['internal_weight']
                node['count'] = tree['internal_count']
            else:
                node['value'] = tree['leaf_value']
                if not _is_single_node_tree(tree):
                    node['weight'] = tree['leaf_weight']
                    node['count'] = tree['leaf_count']

            return node

        def tree_dict_to_node_list(tree, node_depth=1, tree_index=None,
                                   feature_names=None, parent_node=None):

            node = create_node_record(tree,
                                      node_depth=node_depth,
                                      tree_index=tree_index,
                                      feature_names=feature_names,
                                      parent_node=parent_node)

            res = [node]

            if _is_split_node(tree):
                # traverse the next level of the tree
                children = ['left_child', 'right_child']
                for child in children:
                    subtree_list = tree_dict_to_node_list(
                        tree[child],
                        node_depth=node_depth + 1,
                        tree_index=tree_index,
                        feature_names=feature_names,
                        parent_node=node['node_index'])
                    # In tree format, "subtree_list" is a list of node records (dicts),
                    # and we add node to the list.
                    res.extend(subtree_list)
            return res

        model_dict = self.dump_model()
        feature_names = model_dict['feature_names']
        model_list = []
        for tree in model_dict['tree_info']:
            model_list.extend(tree_dict_to_node_list(tree['tree_structure'],
                                                     tree_index=tree['tree_index'],
                                                     feature_names=feature_names))

        return DataFrame(model_list, columns=model_list[0].keys())

wxchan's avatar
wxchan committed
2247
    def set_train_data_name(self, name):
2248
2249
2250
2251
        """Set the name to the training Dataset.

        Parameters
        ----------
Nikita Titov's avatar
Nikita Titov committed
2252
2253
2254
2255
2256
2257
2258
        name : string
            Name for the training Dataset.

        Returns
        -------
        self : Booster
            Booster with set training Dataset name.
2259
        """
2260
        self._train_data_name = name
Nikita Titov's avatar
Nikita Titov committed
2261
        return self
wxchan's avatar
wxchan committed
2262
2263

    def add_valid(self, data, name):
2264
        """Add validation data.
wxchan's avatar
wxchan committed
2265
2266
2267
2268

        Parameters
        ----------
        data : Dataset
2269
2270
2271
            Validation data.
        name : string
            Name of validation data.
Nikita Titov's avatar
Nikita Titov committed
2272
2273
2274
2275
2276

        Returns
        -------
        self : Booster
            Booster with set validation data.
wxchan's avatar
wxchan committed
2277
        """
Guolin Ke's avatar
Guolin Ke committed
2278
        if not isinstance(data, Dataset):
2279
2280
            raise TypeError('Validation data should be Dataset instance, met {}'
                            .format(type(data).__name__))
Guolin Ke's avatar
Guolin Ke committed
2281
        if data._predictor is not self.__init_predictor:
2282
2283
            raise LightGBMError("Add validation data failed, "
                                "you should use same predictor for these data")
wxchan's avatar
wxchan committed
2284
2285
        _safe_call(_LIB.LGBM_BoosterAddValidData(
            self.handle,
wxchan's avatar
wxchan committed
2286
            data.construct().handle))
wxchan's avatar
wxchan committed
2287
2288
2289
2290
2291
        self.valid_sets.append(data)
        self.name_valid_sets.append(name)
        self.__num_dataset += 1
        self.__inner_predict_buffer.append(None)
        self.__is_predicted_cur_iter.append(False)
Nikita Titov's avatar
Nikita Titov committed
2292
        return self
wxchan's avatar
wxchan committed
2293
2294

    def reset_parameter(self, params):
2295
        """Reset parameters of Booster.
wxchan's avatar
wxchan committed
2296
2297
2298
2299

        Parameters
        ----------
        params : dict
2300
            New parameters for Booster.
Nikita Titov's avatar
Nikita Titov committed
2301
2302
2303
2304
2305

        Returns
        -------
        self : Booster
            Booster with new parameters.
wxchan's avatar
wxchan committed
2306
2307
2308
2309
2310
2311
        """
        params_str = param_dict_to_str(params)
        if params_str:
            _safe_call(_LIB.LGBM_BoosterResetParameter(
                self.handle,
                c_str(params_str)))
Guolin Ke's avatar
Guolin Ke committed
2312
        self.params.update(params)
Nikita Titov's avatar
Nikita Titov committed
2313
        return self
wxchan's avatar
wxchan committed
2314
2315

    def update(self, train_set=None, fobj=None):
Nikita Titov's avatar
Nikita Titov committed
2316
        """Update Booster for one iteration.
2317

wxchan's avatar
wxchan committed
2318
2319
        Parameters
        ----------
2320
2321
2322
2323
        train_set : Dataset or None, optional (default=None)
            Training data.
            If None, last training data is used.
        fobj : callable or None, optional (default=None)
wxchan's avatar
wxchan committed
2324
            Customized objective function.
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
            Should accept two parameters: preds, train_data,
            and return (grad, hess).

                preds : list or numpy 1-D array
                    The predicted values.
                train_data : Dataset
                    The training dataset.
                grad : list or numpy 1-D array
                    The value of the first order derivative (gradient) for each sample point.
                hess : list or numpy 1-D array
                    The value of the second order derivative (Hessian) for each sample point.
wxchan's avatar
wxchan committed
2336

2337
            For binary task, the preds is probability of positive class (or margin in case of specified ``fobj``).
2338
2339
            For multi-class task, the preds is group by class_id first, then group by row_id.
            If you want to get i-th row preds in j-th class, the access way is score[j * num_data + i]
2340
2341
            and you should group grad and hess in this way as well.

wxchan's avatar
wxchan committed
2342
2343
        Returns
        -------
2344
2345
        is_finished : bool
            Whether the update was successfully finished.
wxchan's avatar
wxchan committed
2346
        """
2347
        # need reset training data
2348
2349
2350
2351
2352
2353
        if train_set is None and self.train_set_version != self.train_set.version:
            train_set = self.train_set
            is_the_same_train_set = False
        else:
            is_the_same_train_set = train_set is self.train_set and self.train_set_version == train_set.version
        if train_set is not None and not is_the_same_train_set:
Guolin Ke's avatar
Guolin Ke committed
2354
            if not isinstance(train_set, Dataset):
2355
2356
                raise TypeError('Training data should be Dataset instance, met {}'
                                .format(type(train_set).__name__))
Guolin Ke's avatar
Guolin Ke committed
2357
            if train_set._predictor is not self.__init_predictor:
2358
2359
                raise LightGBMError("Replace training data failed, "
                                    "you should use same predictor for these data")
wxchan's avatar
wxchan committed
2360
2361
2362
            self.train_set = train_set
            _safe_call(_LIB.LGBM_BoosterResetTrainingData(
                self.handle,
wxchan's avatar
wxchan committed
2363
                self.train_set.construct().handle))
wxchan's avatar
wxchan committed
2364
            self.__inner_predict_buffer[0] = None
2365
            self.train_set_version = self.train_set.version
wxchan's avatar
wxchan committed
2366
2367
        is_finished = ctypes.c_int(0)
        if fobj is None:
2368
            if self.__set_objective_to_none:
2369
                raise LightGBMError('Cannot update due to null objective function.')
wxchan's avatar
wxchan committed
2370
2371
2372
            _safe_call(_LIB.LGBM_BoosterUpdateOneIter(
                self.handle,
                ctypes.byref(is_finished)))
wxchan's avatar
wxchan committed
2373
            self.__is_predicted_cur_iter = [False for _ in range_(self.__num_dataset)]
wxchan's avatar
wxchan committed
2374
2375
            return is_finished.value == 1
        else:
2376
            if not self.__set_objective_to_none:
Nikita Titov's avatar
Nikita Titov committed
2377
                self.reset_parameter({"objective": "none"}).__set_objective_to_none = True
wxchan's avatar
wxchan committed
2378
2379
2380
2381
            grad, hess = fobj(self.__inner_predict(0), self.train_set)
            return self.__boost(grad, hess)

    def __boost(self, grad, hess):
2382
        """Boost Booster for one iteration with customized gradient statistics.
Nikita Titov's avatar
Nikita Titov committed
2383

Nikita Titov's avatar
Nikita Titov committed
2384
2385
        .. note::

2386
            For binary task, the score is probability of positive class (or margin in case of custom objective).
Nikita Titov's avatar
Nikita Titov committed
2387
2388
2389
            For multi-class task, the score is group by class_id first, then group by row_id.
            If you want to get i-th row score in j-th class, the access way is score[j * num_data + i]
            and you should group grad and hess in this way as well.
2390

wxchan's avatar
wxchan committed
2391
2392
        Parameters
        ----------
2393
        grad : list or numpy 1-D array
Nikita Titov's avatar
Nikita Titov committed
2394
            The first order derivative (gradient).
2395
        hess : list or numpy 1-D array
Nikita Titov's avatar
Nikita Titov committed
2396
            The second order derivative (Hessian).
wxchan's avatar
wxchan committed
2397
2398
2399

        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
2400
2401
        is_finished : bool
            Whether the boost was successfully finished.
wxchan's avatar
wxchan committed
2402
        """
2403
2404
        grad = list_to_1d_numpy(grad, name='gradient')
        hess = list_to_1d_numpy(hess, name='hessian')
2405
2406
        assert grad.flags.c_contiguous
        assert hess.flags.c_contiguous
wxchan's avatar
wxchan committed
2407
        if len(grad) != len(hess):
2408
2409
            raise ValueError("Lengths of gradient({}) and hessian({}) don't match"
                             .format(len(grad), len(hess)))
wxchan's avatar
wxchan committed
2410
2411
2412
2413
2414
2415
        is_finished = ctypes.c_int(0)
        _safe_call(_LIB.LGBM_BoosterUpdateOneIterCustom(
            self.handle,
            grad.ctypes.data_as(ctypes.POINTER(ctypes.c_float)),
            hess.ctypes.data_as(ctypes.POINTER(ctypes.c_float)),
            ctypes.byref(is_finished)))
wxchan's avatar
wxchan committed
2416
        self.__is_predicted_cur_iter = [False for _ in range_(self.__num_dataset)]
wxchan's avatar
wxchan committed
2417
2418
2419
        return is_finished.value == 1

    def rollback_one_iter(self):
Nikita Titov's avatar
Nikita Titov committed
2420
2421
2422
2423
2424
2425
2426
        """Rollback one iteration.

        Returns
        -------
        self : Booster
            Booster with rolled back one iteration.
        """
wxchan's avatar
wxchan committed
2427
2428
        _safe_call(_LIB.LGBM_BoosterRollbackOneIter(
            self.handle))
wxchan's avatar
wxchan committed
2429
        self.__is_predicted_cur_iter = [False for _ in range_(self.__num_dataset)]
Nikita Titov's avatar
Nikita Titov committed
2430
        return self
wxchan's avatar
wxchan committed
2431
2432

    def current_iteration(self):
2433
2434
2435
2436
2437
2438
2439
        """Get the index of the current iteration.

        Returns
        -------
        cur_iter : int
            The index of the current iteration.
        """
Guolin Ke's avatar
Guolin Ke committed
2440
        out_cur_iter = ctypes.c_int(0)
wxchan's avatar
wxchan committed
2441
2442
2443
2444
2445
        _safe_call(_LIB.LGBM_BoosterGetCurrentIteration(
            self.handle,
            ctypes.byref(out_cur_iter)))
        return out_cur_iter.value

2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
    def num_model_per_iteration(self):
        """Get number of models per iteration.

        Returns
        -------
        model_per_iter : int
            The number of models per iteration.
        """
        model_per_iter = ctypes.c_int(0)
        _safe_call(_LIB.LGBM_BoosterNumModelPerIteration(
            self.handle,
            ctypes.byref(model_per_iter)))
        return model_per_iter.value

    def num_trees(self):
        """Get number of weak sub-models.

        Returns
        -------
        num_trees : int
            The number of weak sub-models.
        """
        num_trees = ctypes.c_int(0)
        _safe_call(_LIB.LGBM_BoosterNumberOfTotalModel(
            self.handle,
            ctypes.byref(num_trees)))
        return num_trees.value

2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
    def upper_bound(self):
        """Get upper bound value of a model.

        Returns
        -------
        upper_bound : double
            Upper bound value of the model.
        """
        ret = ctypes.c_double(0)
        _safe_call(_LIB.LGBM_BoosterGetUpperBoundValue(
            self.handle,
            ctypes.byref(ret)))
        return ret.value

    def lower_bound(self):
        """Get lower bound value of a model.

        Returns
        -------
        lower_bound : double
            Lower bound value of the model.
        """
        ret = ctypes.c_double(0)
        _safe_call(_LIB.LGBM_BoosterGetLowerBoundValue(
            self.handle,
            ctypes.byref(ret)))
        return ret.value

wxchan's avatar
wxchan committed
2502
    def eval(self, data, name, feval=None):
2503
        """Evaluate for data.
wxchan's avatar
wxchan committed
2504
2505
2506

        Parameters
        ----------
2507
2508
2509
2510
2511
        data : Dataset
            Data for the evaluating.
        name : string
            Name of the data.
        feval : callable or None, optional (default=None)
2512
            Customized evaluation function.
2513
            Should accept two parameters: preds, eval_data,
2514
            and return (eval_name, eval_result, is_higher_better) or list of such tuples.
2515
2516
2517
2518
2519
2520

                preds : list or numpy 1-D array
                    The predicted values.
                eval_data : Dataset
                    The evaluation dataset.
                eval_name : string
2521
                    The name of evaluation function (without whitespaces).
2522
2523
2524
2525
2526
                eval_result : float
                    The eval result.
                is_higher_better : bool
                    Is eval result higher better, e.g. AUC is ``is_higher_better``.

2527
            For binary task, the preds is probability of positive class (or margin in case of specified ``fobj``).
2528
2529
            For multi-class task, the preds is group by class_id first, then group by row_id.
            If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
2530

wxchan's avatar
wxchan committed
2531
2532
        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
2533
        result : list
2534
            List with evaluation results.
wxchan's avatar
wxchan committed
2535
        """
Guolin Ke's avatar
Guolin Ke committed
2536
2537
        if not isinstance(data, Dataset):
            raise TypeError("Can only eval for Dataset instance")
wxchan's avatar
wxchan committed
2538
2539
2540
2541
        data_idx = -1
        if data is self.train_set:
            data_idx = 0
        else:
wxchan's avatar
wxchan committed
2542
            for i in range_(len(self.valid_sets)):
wxchan's avatar
wxchan committed
2543
2544
2545
                if data is self.valid_sets[i]:
                    data_idx = i + 1
                    break
2546
        # need to push new valid data
wxchan's avatar
wxchan committed
2547
2548
2549
2550
2551
2552
2553
        if data_idx == -1:
            self.add_valid(data, name)
            data_idx = self.__num_dataset - 1

        return self.__inner_eval(name, data_idx, feval)

    def eval_train(self, feval=None):
2554
        """Evaluate for training data.
wxchan's avatar
wxchan committed
2555
2556
2557

        Parameters
        ----------
2558
        feval : callable or None, optional (default=None)
2559
            Customized evaluation function.
2560
2561
            Should accept two parameters: preds, train_data,
            and return (eval_name, eval_result, is_higher_better) or list of such tuples.
2562
2563
2564
2565
2566
2567

                preds : list or numpy 1-D array
                    The predicted values.
                train_data : Dataset
                    The training dataset.
                eval_name : string
2568
                    The name of evaluation function (without whitespaces).
2569
2570
2571
2572
2573
                eval_result : float
                    The eval result.
                is_higher_better : bool
                    Is eval result higher better, e.g. AUC is ``is_higher_better``.

2574
            For binary task, the preds is probability of positive class (or margin in case of specified ``fobj``).
2575
2576
            For multi-class task, the preds is group by class_id first, then group by row_id.
            If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
wxchan's avatar
wxchan committed
2577
2578
2579

        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
2580
        result : list
2581
            List with evaluation results.
wxchan's avatar
wxchan committed
2582
        """
2583
        return self.__inner_eval(self._train_data_name, 0, feval)
wxchan's avatar
wxchan committed
2584
2585

    def eval_valid(self, feval=None):
2586
        """Evaluate for validation data.
wxchan's avatar
wxchan committed
2587
2588
2589

        Parameters
        ----------
2590
        feval : callable or None, optional (default=None)
2591
            Customized evaluation function.
2592
            Should accept two parameters: preds, valid_data,
2593
            and return (eval_name, eval_result, is_higher_better) or list of such tuples.
2594
2595
2596
2597
2598
2599

                preds : list or numpy 1-D array
                    The predicted values.
                valid_data : Dataset
                    The validation dataset.
                eval_name : string
2600
                    The name of evaluation function (without whitespaces).
2601
2602
2603
2604
2605
                eval_result : float
                    The eval result.
                is_higher_better : bool
                    Is eval result higher better, e.g. AUC is ``is_higher_better``.

2606
            For binary task, the preds is probability of positive class (or margin in case of specified ``fobj``).
2607
2608
            For multi-class task, the preds is group by class_id first, then group by row_id.
            If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
wxchan's avatar
wxchan committed
2609
2610
2611

        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
2612
        result : list
2613
            List with evaluation results.
wxchan's avatar
wxchan committed
2614
        """
wxchan's avatar
wxchan committed
2615
        return [item for i in range_(1, self.__num_dataset)
wxchan's avatar
wxchan committed
2616
                for item in self.__inner_eval(self.name_valid_sets[i - 1], i, feval)]
wxchan's avatar
wxchan committed
2617

2618
    def save_model(self, filename, num_iteration=None, start_iteration=0, importance_type='split'):
2619
        """Save Booster to file.
wxchan's avatar
wxchan committed
2620
2621
2622

        Parameters
        ----------
2623
2624
        filename : string
            Filename to save Booster.
2625
2626
2627
2628
        num_iteration : int or None, optional (default=None)
            Index of the iteration that should be saved.
            If None, if the best iteration exists, it is saved; otherwise, all iterations are saved.
            If <= 0, all iterations are saved.
Nikita Titov's avatar
Nikita Titov committed
2629
        start_iteration : int, optional (default=0)
2630
            Start index of the iteration that should be saved.
2631
2632
2633
2634
        importance_type : string, optional (default="split")
            What type of feature importance should be saved.
            If "split", result contains numbers of times the feature is used in a model.
            If "gain", result contains total gains of splits which use the feature.
Nikita Titov's avatar
Nikita Titov committed
2635
2636
2637
2638
2639

        Returns
        -------
        self : Booster
            Returns self.
wxchan's avatar
wxchan committed
2640
        """
2641
        if num_iteration is None:
2642
            num_iteration = self.best_iteration
2643
        importance_type_int = FEATURE_IMPORTANCE_TYPE_MAPPER[importance_type]
wxchan's avatar
wxchan committed
2644
2645
        _safe_call(_LIB.LGBM_BoosterSaveModel(
            self.handle,
2646
            ctypes.c_int(start_iteration),
Guolin Ke's avatar
Guolin Ke committed
2647
            ctypes.c_int(num_iteration),
2648
            ctypes.c_int(importance_type_int),
wxchan's avatar
wxchan committed
2649
            c_str(filename)))
2650
        _dump_pandas_categorical(self.pandas_categorical, filename)
Nikita Titov's avatar
Nikita Titov committed
2651
        return self
wxchan's avatar
wxchan committed
2652

2653
    def shuffle_models(self, start_iteration=0, end_iteration=-1):
2654
        """Shuffle models.
Nikita Titov's avatar
Nikita Titov committed
2655

2656
2657
2658
        Parameters
        ----------
        start_iteration : int, optional (default=0)
2659
            The first iteration that will be shuffled.
2660
2661
        end_iteration : int, optional (default=-1)
            The last iteration that will be shuffled.
2662
            If <= 0, means the last available iteration.
2663

Nikita Titov's avatar
Nikita Titov committed
2664
2665
2666
2667
        Returns
        -------
        self : Booster
            Booster with shuffled models.
2668
        """
2669
2670
        _safe_call(_LIB.LGBM_BoosterShuffleModels(
            self.handle,
Guolin Ke's avatar
Guolin Ke committed
2671
2672
            ctypes.c_int(start_iteration),
            ctypes.c_int(end_iteration)))
Nikita Titov's avatar
Nikita Titov committed
2673
        return self
2674
2675
2676
2677
2678
2679

    def model_from_string(self, model_str, verbose=True):
        """Load Booster from a string.

        Parameters
        ----------
Nikita Titov's avatar
Nikita Titov committed
2680
        model_str : string
2681
            Model will be loaded from this string.
Nikita Titov's avatar
Nikita Titov committed
2682
2683
        verbose : bool, optional (default=True)
            Whether to print messages while loading model.
2684
2685
2686

        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
2687
        self : Booster
2688
2689
            Loaded Booster object.
        """
2690
2691
2692
2693
        if self.handle is not None:
            _safe_call(_LIB.LGBM_BoosterFree(self.handle))
        self._free_buffer()
        self.handle = ctypes.c_void_p()
2694
2695
2696
2697
2698
2699
2700
2701
2702
        out_num_iterations = ctypes.c_int(0)
        _safe_call(_LIB.LGBM_BoosterLoadModelFromString(
            c_str(model_str),
            ctypes.byref(out_num_iterations),
            ctypes.byref(self.handle)))
        out_num_class = ctypes.c_int(0)
        _safe_call(_LIB.LGBM_BoosterGetNumClasses(
            self.handle,
            ctypes.byref(out_num_class)))
2703
        if verbose:
Nikita Titov's avatar
Nikita Titov committed
2704
            print('Finished loading model, total used %d iterations' % int(out_num_iterations.value))
2705
        self.__num_class = out_num_class.value
2706
        self.pandas_categorical = _load_pandas_categorical(model_str=model_str)
2707
2708
        return self

2709
    def model_to_string(self, num_iteration=None, start_iteration=0, importance_type='split'):
2710
        """Save Booster to string.
2711

2712
2713
2714
2715
2716
2717
        Parameters
        ----------
        num_iteration : int or None, optional (default=None)
            Index of the iteration that should be saved.
            If None, if the best iteration exists, it is saved; otherwise, all iterations are saved.
            If <= 0, all iterations are saved.
Nikita Titov's avatar
Nikita Titov committed
2718
        start_iteration : int, optional (default=0)
2719
            Start index of the iteration that should be saved.
2720
2721
2722
2723
        importance_type : string, optional (default="split")
            What type of feature importance should be saved.
            If "split", result contains numbers of times the feature is used in a model.
            If "gain", result contains total gains of splits which use the feature.
2724
2725
2726

        Returns
        -------
Nikita Titov's avatar
Nikita Titov committed
2727
        str_repr : string
2728
2729
            String representation of Booster.
        """
2730
        if num_iteration is None:
2731
            num_iteration = self.best_iteration
2732
        importance_type_int = FEATURE_IMPORTANCE_TYPE_MAPPER[importance_type]
2733
        buffer_len = 1 << 20
2734
        tmp_out_len = ctypes.c_int64(0)
2735
2736
2737
2738
        string_buffer = ctypes.create_string_buffer(buffer_len)
        ptr_string_buffer = ctypes.c_char_p(*[ctypes.addressof(string_buffer)])
        _safe_call(_LIB.LGBM_BoosterSaveModelToString(
            self.handle,
2739
            ctypes.c_int(start_iteration),
2740
            ctypes.c_int(num_iteration),
2741
            ctypes.c_int(importance_type_int),
2742
            ctypes.c_int64(buffer_len),
2743
2744
2745
            ctypes.byref(tmp_out_len),
            ptr_string_buffer))
        actual_len = tmp_out_len.value
2746
        # if buffer length is not long enough, re-allocate a buffer
2747
2748
2749
2750
2751
        if actual_len > buffer_len:
            string_buffer = ctypes.create_string_buffer(actual_len)
            ptr_string_buffer = ctypes.c_char_p(*[ctypes.addressof(string_buffer)])
            _safe_call(_LIB.LGBM_BoosterSaveModelToString(
                self.handle,
2752
                ctypes.c_int(start_iteration),
2753
                ctypes.c_int(num_iteration),
2754
                ctypes.c_int(importance_type_int),
2755
                ctypes.c_int64(actual_len),
2756
2757
                ctypes.byref(tmp_out_len),
                ptr_string_buffer))
2758
        ret = string_buffer.value.decode('utf-8')
2759
2760
        ret += _dump_pandas_categorical(self.pandas_categorical)
        return ret
2761

2762
    def dump_model(self, num_iteration=None, start_iteration=0, importance_type='split'):
Nikita Titov's avatar
Nikita Titov committed
2763
        """Dump Booster to JSON format.
wxchan's avatar
wxchan committed
2764

2765
2766
        Parameters
        ----------
2767
2768
2769
2770
        num_iteration : int or None, optional (default=None)
            Index of the iteration that should be dumped.
            If None, if the best iteration exists, it is dumped; otherwise, all iterations are dumped.
            If <= 0, all iterations are dumped.
Nikita Titov's avatar
Nikita Titov committed
2771
        start_iteration : int, optional (default=0)
2772
            Start index of the iteration that should be dumped.
2773
2774
2775
2776
        importance_type : string, optional (default="split")
            What type of feature importance should be dumped.
            If "split", result contains numbers of times the feature is used in a model.
            If "gain", result contains total gains of splits which use the feature.
2777

wxchan's avatar
wxchan committed
2778
2779
        Returns
        -------
2780
        json_repr : dict
Nikita Titov's avatar
Nikita Titov committed
2781
            JSON format of Booster.
wxchan's avatar
wxchan committed
2782
        """
2783
        if num_iteration is None:
2784
            num_iteration = self.best_iteration
2785
        importance_type_int = FEATURE_IMPORTANCE_TYPE_MAPPER[importance_type]
wxchan's avatar
wxchan committed
2786
        buffer_len = 1 << 20
2787
        tmp_out_len = ctypes.c_int64(0)
wxchan's avatar
wxchan committed
2788
2789
2790
2791
        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,
2792
            ctypes.c_int(start_iteration),
Guolin Ke's avatar
Guolin Ke committed
2793
            ctypes.c_int(num_iteration),
2794
            ctypes.c_int(importance_type_int),
2795
            ctypes.c_int64(buffer_len),
wxchan's avatar
wxchan committed
2796
            ctypes.byref(tmp_out_len),
Guolin Ke's avatar
Guolin Ke committed
2797
            ptr_string_buffer))
wxchan's avatar
wxchan committed
2798
        actual_len = tmp_out_len.value
2799
        # if buffer length is not long enough, reallocate a buffer
wxchan's avatar
wxchan committed
2800
2801
2802
2803
2804
        if actual_len > buffer_len:
            string_buffer = ctypes.create_string_buffer(actual_len)
            ptr_string_buffer = ctypes.c_char_p(*[ctypes.addressof(string_buffer)])
            _safe_call(_LIB.LGBM_BoosterDumpModel(
                self.handle,
2805
                ctypes.c_int(start_iteration),
Guolin Ke's avatar
Guolin Ke committed
2806
                ctypes.c_int(num_iteration),
2807
                ctypes.c_int(importance_type_int),
2808
                ctypes.c_int64(actual_len),
wxchan's avatar
wxchan committed
2809
                ctypes.byref(tmp_out_len),
Guolin Ke's avatar
Guolin Ke committed
2810
                ptr_string_buffer))
2811
        ret = json.loads(string_buffer.value.decode('utf-8'))
2812
2813
2814
        ret['pandas_categorical'] = json.loads(json.dumps(self.pandas_categorical,
                                                          default=json_default_with_numpy))
        return ret
wxchan's avatar
wxchan committed
2815

2816
    def predict(self, data, start_iteration=0, num_iteration=None,
2817
                raw_score=False, pred_leaf=False, pred_contrib=False,
2818
                data_has_header=False, is_reshape=True, **kwargs):
2819
        """Make a prediction.
wxchan's avatar
wxchan committed
2820
2821
2822

        Parameters
        ----------
2823
        data : string, numpy array, pandas DataFrame, H2O DataTable's Frame or scipy.sparse
2824
2825
            Data source for prediction.
            If string, it represents the path to txt file.
2826
        start_iteration : int, optional (default=0)
2827
            Start index of the iteration to predict.
2828
            If <= 0, starts from the first iteration.
2829
        num_iteration : int or None, optional (default=None)
2830
2831
2832
2833
            Total number of iterations used in the prediction.
            If None, if the best iteration exists and start_iteration <= 0, the best iteration is used;
            otherwise, all iterations from ``start_iteration`` are used (no limits).
            If <= 0, all iterations from ``start_iteration`` are used (no limits).
2834
2835
2836
2837
        raw_score : bool, optional (default=False)
            Whether to predict raw scores.
        pred_leaf : bool, optional (default=False)
            Whether to predict leaf index.
2838
2839
        pred_contrib : bool, optional (default=False)
            Whether to predict feature contributions.
2840

Nikita Titov's avatar
Nikita Titov committed
2841
2842
2843
2844
2845
2846
2847
            .. note::

                If you want to get more explanations for your model's predictions using SHAP values,
                like SHAP interaction values,
                you can install the shap package (https://github.com/slundberg/shap).
                Note that unlike the shap package, with ``pred_contrib`` we return a matrix with an extra
                column, where the last column is the expected value.
2848

2849
2850
2851
2852
2853
        data_has_header : bool, optional (default=False)
            Whether the data has header.
            Used only if data is string.
        is_reshape : bool, optional (default=True)
            If True, result is reshaped to [nrow, ncol].
2854
2855
        **kwargs
            Other parameters for the prediction.
wxchan's avatar
wxchan committed
2856
2857
2858

        Returns
        -------
2859
        result : numpy array, scipy.sparse or list of scipy.sparse
2860
            Prediction result.
2861
            Can be sparse or a list of sparse objects (each element represents predictions for one class) for feature contributions (when ``pred_contrib=True``).
wxchan's avatar
wxchan committed
2862
        """
2863
        predictor = self._to_predictor(copy.deepcopy(kwargs))
2864
        if num_iteration is None:
2865
            if start_iteration <= 0:
2866
2867
2868
2869
                num_iteration = self.best_iteration
            else:
                num_iteration = -1
        return predictor.predict(data, start_iteration, num_iteration,
2870
2871
                                 raw_score, pred_leaf, pred_contrib,
                                 data_has_header, is_reshape)
wxchan's avatar
wxchan committed
2872

2873
    def refit(self, data, label, decay_rate=0.9, **kwargs):
Guolin Ke's avatar
Guolin Ke committed
2874
2875
2876
2877
        """Refit the existing Booster by new data.

        Parameters
        ----------
2878
        data : string, numpy array, pandas DataFrame, H2O DataTable's Frame or scipy.sparse
Guolin Ke's avatar
Guolin Ke committed
2879
2880
            Data source for refit.
            If string, it represents the path to txt file.
2881
        label : list, numpy 1-D array or pandas Series / one-column DataFrame
Guolin Ke's avatar
Guolin Ke committed
2882
2883
            Label for refit.
        decay_rate : float, optional (default=0.9)
2884
2885
            Decay rate of refit,
            will use ``leaf_output = decay_rate * old_leaf_output + (1.0 - decay_rate) * new_leaf_output`` to refit trees.
2886
2887
        **kwargs
            Other parameters for refit.
2888
            These parameters will be passed to ``predict`` method.
Guolin Ke's avatar
Guolin Ke committed
2889
2890
2891
2892
2893
2894

        Returns
        -------
        result : Booster
            Refitted Booster.
        """
2895
2896
        if self.__set_objective_to_none:
            raise LightGBMError('Cannot refit due to null objective function.')
2897
        predictor = self._to_predictor(copy.deepcopy(kwargs))
2898
        leaf_preds = predictor.predict(data, -1, pred_leaf=True)
2899
        nrow, ncol = leaf_preds.shape
2900
        train_set = Dataset(data, label, silent=True)
2901
2902
        new_params = copy.deepcopy(self.params)
        new_params['refit_decay_rate'] = decay_rate
2903
        new_booster = Booster(new_params, train_set)
Guolin Ke's avatar
Guolin Ke committed
2904
2905
2906
2907
2908
        # Copy models
        _safe_call(_LIB.LGBM_BoosterMerge(
            new_booster.handle,
            predictor.handle))
        leaf_preds = leaf_preds.reshape(-1)
2909
        ptr_data, _, _ = c_int_array(leaf_preds)
Guolin Ke's avatar
Guolin Ke committed
2910
2911
2912
2913
2914
        _safe_call(_LIB.LGBM_BoosterRefit(
            new_booster.handle,
            ptr_data,
            ctypes.c_int(nrow),
            ctypes.c_int(ncol)))
2915
2916
        new_booster.network = self.network
        new_booster.__attr = self.__attr.copy()
Guolin Ke's avatar
Guolin Ke committed
2917
2918
        return new_booster

2919
    def get_leaf_output(self, tree_id, leaf_id):
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
        """Get the output of a leaf.

        Parameters
        ----------
        tree_id : int
            The index of the tree.
        leaf_id : int
            The index of the leaf in the tree.

        Returns
        -------
        result : float
            The output of the leaf.
        """
2934
2935
2936
2937
2938
2939
2940
2941
        ret = ctypes.c_double(0)
        _safe_call(_LIB.LGBM_BoosterGetLeafValue(
            self.handle,
            ctypes.c_int(tree_id),
            ctypes.c_int(leaf_id),
            ctypes.byref(ret)))
        return ret.value

2942
    def _to_predictor(self, pred_parameter=None):
2943
        """Convert to predictor."""
2944
        predictor = _InnerPredictor(booster_handle=self.handle, pred_parameter=pred_parameter)
2945
        predictor.pandas_categorical = self.pandas_categorical
wxchan's avatar
wxchan committed
2946
2947
        return predictor

2948
    def num_feature(self):
2949
2950
2951
2952
2953
2954
2955
        """Get number of features.

        Returns
        -------
        num_feature : int
            The number of features.
        """
2956
2957
2958
2959
2960
2961
        out_num_feature = ctypes.c_int(0)
        _safe_call(_LIB.LGBM_BoosterGetNumFeature(
            self.handle,
            ctypes.byref(out_num_feature)))
        return out_num_feature.value

wxchan's avatar
wxchan committed
2962
    def feature_name(self):
2963
        """Get names of features.
wxchan's avatar
wxchan committed
2964
2965
2966

        Returns
        -------
2967
2968
        result : list
            List with names of features.
wxchan's avatar
wxchan committed
2969
        """
2970
        num_feature = self.num_feature()
2971
        # Get name of features
wxchan's avatar
wxchan committed
2972
        tmp_out_len = ctypes.c_int(0)
2973
2974
2975
        reserved_string_buffer_size = 255
        required_string_buffer_size = ctypes.c_size_t(0)
        string_buffers = [ctypes.create_string_buffer(reserved_string_buffer_size) for i in range_(num_feature)]
wxchan's avatar
wxchan committed
2976
2977
2978
        ptr_string_buffers = (ctypes.c_char_p * num_feature)(*map(ctypes.addressof, string_buffers))
        _safe_call(_LIB.LGBM_BoosterGetFeatureNames(
            self.handle,
2979
            ctypes.c_int(num_feature),
wxchan's avatar
wxchan committed
2980
            ctypes.byref(tmp_out_len),
2981
            ctypes.c_size_t(reserved_string_buffer_size),
2982
            ctypes.byref(required_string_buffer_size),
wxchan's avatar
wxchan committed
2983
2984
2985
            ptr_string_buffers))
        if num_feature != tmp_out_len.value:
            raise ValueError("Length of feature names doesn't equal with num_feature")
2986
2987
2988
2989
2990
        if reserved_string_buffer_size < required_string_buffer_size.value:
            raise BufferError(
                "Allocated feature name buffer size ({}) was inferior to the needed size ({})."
                .format(reserved_string_buffer_size, required_string_buffer_size.value)
            )
2991
        return [string_buffers[i].value.decode('utf-8') for i in range_(num_feature)]
wxchan's avatar
wxchan committed
2992

2993
    def feature_importance(self, importance_type='split', iteration=None):
2994
        """Get feature importances.
2995

2996
2997
        Parameters
        ----------
2998
2999
3000
3001
        importance_type : string, optional (default="split")
            How the importance is calculated.
            If "split", result contains numbers of times the feature is used in a model.
            If "gain", result contains total gains of splits which use the feature.
3002
3003
3004
3005
        iteration : int or None, optional (default=None)
            Limit number of iterations in the feature importance calculation.
            If None, if the best iteration exists, it is used; otherwise, all trees are used.
            If <= 0, all trees are used (no limits).
3006

3007
3008
        Returns
        -------
3009
3010
        result : numpy array
            Array with feature importances.
3011
        """
3012
3013
        if iteration is None:
            iteration = self.best_iteration
3014
        importance_type_int = FEATURE_IMPORTANCE_TYPE_MAPPER[importance_type]
Nikita Titov's avatar
Nikita Titov committed
3015
        result = np.zeros(self.num_feature(), dtype=np.float64)
3016
3017
3018
3019
3020
3021
        _safe_call(_LIB.LGBM_BoosterFeatureImportance(
            self.handle,
            ctypes.c_int(iteration),
            ctypes.c_int(importance_type_int),
            result.ctypes.data_as(ctypes.POINTER(ctypes.c_double))))
        if importance_type_int == 0:
3022
            return result.astype(np.int32)
3023
3024
        else:
            return result
3025

3026
3027
3028
3029
3030
3031
3032
3033
3034
    def get_split_value_histogram(self, feature, bins=None, xgboost_style=False):
        """Get split value histogram for the specified feature.

        Parameters
        ----------
        feature : int or string
            The feature name or index the histogram is calculated for.
            If int, interpreted as index.
            If string, interpreted as name.
3035

Nikita Titov's avatar
Nikita Titov committed
3036
3037
3038
            .. warning::

                Categorical features are not supported.
3039

3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
        bins : int, string or None, optional (default=None)
            The maximum number of bins.
            If None, or int and > number of unique split values and ``xgboost_style=True``,
            the number of bins equals number of unique split values.
            If string, it should be one from the list of the supported values by ``numpy.histogram()`` function.
        xgboost_style : bool, optional (default=False)
            Whether the returned result should be in the same form as it is in XGBoost.
            If False, the returned value is tuple of 2 numpy arrays as it is in ``numpy.histogram()`` function.
            If True, the returned value is matrix, in which the first column is the right edges of non-empty bins
            and the second one is the histogram values.

        Returns
        -------
        result_tuple : tuple of 2 numpy arrays
            If ``xgboost_style=False``, the values of the histogram of used splitting values for the specified feature
            and the bin edges.
        result_array_like : numpy array or pandas DataFrame (if pandas is installed)
            If ``xgboost_style=True``, the histogram of used splitting values for the specified feature.
        """
        def add(root):
            """Recursively add thresholds."""
            if 'split_index' in root:  # non-leaf
                if feature_names is not None and isinstance(feature, string_type):
                    split_feature = feature_names[root['split_feature']]
                else:
                    split_feature = root['split_feature']
                if split_feature == feature:
3067
3068
3069
3070
                    if isinstance(root['threshold'], string_type):
                        raise LightGBMError('Cannot compute split value histogram for the categorical feature')
                    else:
                        values.append(root['threshold'])
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
                add(root['left_child'])
                add(root['right_child'])

        model = self.dump_model()
        feature_names = model.get('feature_names')
        tree_infos = model['tree_info']
        values = []
        for tree_info in tree_infos:
            add(tree_info['tree_structure'])

        if bins is None or isinstance(bins, integer_types) and xgboost_style:
            n_unique = len(np.unique(values))
            bins = max(min(n_unique, bins) if bins is not None else n_unique, 1)
        hist, bin_edges = np.histogram(values, bins=bins)
        if xgboost_style:
            ret = np.column_stack((bin_edges[1:], hist))
            ret = ret[ret[:, 1] > 0]
            if PANDAS_INSTALLED:
                return DataFrame(ret, columns=['SplitValue', 'Count'])
            else:
                return ret
        else:
            return hist, bin_edges

wxchan's avatar
wxchan committed
3095
    def __inner_eval(self, data_name, data_idx, feval=None):
3096
        """Evaluate training or validation data."""
wxchan's avatar
wxchan committed
3097
        if data_idx >= self.__num_dataset:
3098
            raise ValueError("Data_idx should be smaller than number of dataset")
wxchan's avatar
wxchan committed
3099
3100
3101
        self.__get_eval_info()
        ret = []
        if self.__num_inner_eval > 0:
3102
            result = np.zeros(self.__num_inner_eval, dtype=np.float64)
Guolin Ke's avatar
Guolin Ke committed
3103
            tmp_out_len = ctypes.c_int(0)
wxchan's avatar
wxchan committed
3104
3105
            _safe_call(_LIB.LGBM_BoosterGetEval(
                self.handle,
Guolin Ke's avatar
Guolin Ke committed
3106
                ctypes.c_int(data_idx),
wxchan's avatar
wxchan committed
3107
                ctypes.byref(tmp_out_len),
Guolin Ke's avatar
Guolin Ke committed
3108
                result.ctypes.data_as(ctypes.POINTER(ctypes.c_double))))
wxchan's avatar
wxchan committed
3109
            if tmp_out_len.value != self.__num_inner_eval:
3110
                raise ValueError("Wrong length of eval results")
wxchan's avatar
wxchan committed
3111
            for i in range_(self.__num_inner_eval):
3112
3113
                ret.append((data_name, self.__name_inner_eval[i],
                            result[i], self.__higher_better_inner_eval[i]))
3114
3115
        if callable(feval):
            feval = [feval]
wxchan's avatar
wxchan committed
3116
3117
3118
3119
3120
        if feval is not None:
            if data_idx == 0:
                cur_data = self.train_set
            else:
                cur_data = self.valid_sets[data_idx - 1]
3121
3122
3123
3124
3125
3126
3127
3128
3129
            for eval_function in feval:
                if eval_function is None:
                    continue
                feval_ret = eval_function(self.__inner_predict(data_idx), cur_data)
                if isinstance(feval_ret, list):
                    for eval_name, val, is_higher_better in feval_ret:
                        ret.append((data_name, eval_name, val, is_higher_better))
                else:
                    eval_name, val, is_higher_better = feval_ret
wxchan's avatar
wxchan committed
3130
3131
3132
3133
                    ret.append((data_name, eval_name, val, is_higher_better))
        return ret

    def __inner_predict(self, data_idx):
3134
        """Predict for training and validation dataset."""
wxchan's avatar
wxchan committed
3135
        if data_idx >= self.__num_dataset:
3136
            raise ValueError("Data_idx should be smaller than number of dataset")
wxchan's avatar
wxchan committed
3137
3138
3139
3140
3141
        if self.__inner_predict_buffer[data_idx] is None:
            if data_idx == 0:
                n_preds = self.train_set.num_data() * self.__num_class
            else:
                n_preds = self.valid_sets[data_idx - 1].num_data() * self.__num_class
3142
            self.__inner_predict_buffer[data_idx] = np.zeros(n_preds, dtype=np.float64)
3143
        # avoid to predict many time in one iteration
wxchan's avatar
wxchan committed
3144
3145
        if not self.__is_predicted_cur_iter[data_idx]:
            tmp_out_len = ctypes.c_int64(0)
Guolin Ke's avatar
Guolin Ke committed
3146
            data_ptr = self.__inner_predict_buffer[data_idx].ctypes.data_as(ctypes.POINTER(ctypes.c_double))
wxchan's avatar
wxchan committed
3147
3148
            _safe_call(_LIB.LGBM_BoosterGetPredict(
                self.handle,
Guolin Ke's avatar
Guolin Ke committed
3149
                ctypes.c_int(data_idx),
wxchan's avatar
wxchan committed
3150
3151
3152
                ctypes.byref(tmp_out_len),
                data_ptr))
            if tmp_out_len.value != len(self.__inner_predict_buffer[data_idx]):
3153
                raise ValueError("Wrong length of predict results for data %d" % (data_idx))
wxchan's avatar
wxchan committed
3154
3155
3156
3157
            self.__is_predicted_cur_iter[data_idx] = True
        return self.__inner_predict_buffer[data_idx]

    def __get_eval_info(self):
3158
        """Get inner evaluation count and names."""
wxchan's avatar
wxchan committed
3159
3160
        if self.__need_reload_eval_info:
            self.__need_reload_eval_info = False
Guolin Ke's avatar
Guolin Ke committed
3161
            out_num_eval = ctypes.c_int(0)
3162
            # Get num of inner evals
wxchan's avatar
wxchan committed
3163
3164
3165
3166
3167
            _safe_call(_LIB.LGBM_BoosterGetEvalCounts(
                self.handle,
                ctypes.byref(out_num_eval)))
            self.__num_inner_eval = out_num_eval.value
            if self.__num_inner_eval > 0:
3168
                # Get name of evals
Guolin Ke's avatar
Guolin Ke committed
3169
                tmp_out_len = ctypes.c_int(0)
3170
3171
3172
3173
3174
                reserved_string_buffer_size = 255
                required_string_buffer_size = ctypes.c_size_t(0)
                string_buffers = [
                    ctypes.create_string_buffer(reserved_string_buffer_size) for i in range_(self.__num_inner_eval)
                ]
wxchan's avatar
wxchan committed
3175
                ptr_string_buffers = (ctypes.c_char_p * self.__num_inner_eval)(*map(ctypes.addressof, string_buffers))
wxchan's avatar
wxchan committed
3176
3177
                _safe_call(_LIB.LGBM_BoosterGetEvalNames(
                    self.handle,
3178
                    ctypes.c_int(self.__num_inner_eval),
wxchan's avatar
wxchan committed
3179
                    ctypes.byref(tmp_out_len),
3180
                    ctypes.c_size_t(reserved_string_buffer_size),
3181
                    ctypes.byref(required_string_buffer_size),
wxchan's avatar
wxchan committed
3182
3183
                    ptr_string_buffers))
                if self.__num_inner_eval != tmp_out_len.value:
3184
                    raise ValueError("Length of eval names doesn't equal with num_evals")
3185
3186
3187
3188
3189
                if reserved_string_buffer_size < required_string_buffer_size.value:
                    raise BufferError(
                        "Allocated eval name buffer size ({}) was inferior to the needed size ({})."
                        .format(reserved_string_buffer_size, required_string_buffer_size.value)
                    )
3190
                self.__name_inner_eval = \
3191
                    [string_buffers[i].value.decode('utf-8') for i in range_(self.__num_inner_eval)]
3192
                self.__higher_better_inner_eval = \
3193
                    [name.startswith(('auc', 'ndcg@', 'map@')) for name in self.__name_inner_eval]
3194

wxchan's avatar
wxchan committed
3195
    def attr(self, key):
3196
        """Get attribute string from the Booster.
wxchan's avatar
wxchan committed
3197
3198
3199

        Parameters
        ----------
3200
3201
        key : string
            The name of the attribute.
wxchan's avatar
wxchan committed
3202
3203
3204

        Returns
        -------
3205
3206
        value : string or None
            The attribute value.
Nikita Titov's avatar
Nikita Titov committed
3207
            Returns None if attribute does not exist.
wxchan's avatar
wxchan committed
3208
        """
3209
        return self.__attr.get(key, None)
wxchan's avatar
wxchan committed
3210
3211

    def set_attr(self, **kwargs):
3212
        """Set attributes to the Booster.
wxchan's avatar
wxchan committed
3213
3214
3215
3216

        Parameters
        ----------
        **kwargs
3217
3218
            The attributes to set.
            Setting a value to None deletes an attribute.
Nikita Titov's avatar
Nikita Titov committed
3219
3220
3221
3222

        Returns
        -------
        self : Booster
3223
            Booster with set attributes.
wxchan's avatar
wxchan committed
3224
3225
3226
        """
        for key, value in kwargs.items():
            if value is not None:
wxchan's avatar
wxchan committed
3227
                if not isinstance(value, string_type):
Nikita Titov's avatar
Nikita Titov committed
3228
                    raise ValueError("Only string values are accepted")
wxchan's avatar
wxchan committed
3229
3230
3231
                self.__attr[key] = value
            else:
                self.__attr.pop(key, None)
Nikita Titov's avatar
Nikita Titov committed
3232
        return self