MCPcopy Create free account
hub / github.com/antmachineintelligence/mtgbmcode / Dataset

Class Dataset

python-package/lightgbmmt/basic.py:717–1661  ·  view source on GitHub ↗

Dataset in LightGBM.

Source from the content-addressed store, hash-verified

715
716
717class Dataset(object):
718 """Dataset in LightGBM."""
719
720 def __init__(self, data, label=None, reference=None,
721 weight=None, group=None, init_score=None, silent=False,
722 feature_name='auto', categorical_feature='auto', params=None,
723 free_raw_data=True):
724 """Initialize Dataset.
725
726 Parameters
727 ----------
728 data : string, numpy array, pandas DataFrame, H2O DataTable's Frame, scipy.sparse or list of numpy arrays
729 Data source of Dataset.
730 If string, it represents the path to txt file.
731 label : list, numpy 1-D array, pandas Series / one-column DataFrame or None, optional (default=None)
732 Label of the data.
733 reference : Dataset or None, optional (default=None)
734 If this is Dataset for validation, training data should be used as reference.
735 weight : list, numpy 1-D array, pandas Series or None, optional (default=None)
736 Weight for each instance.
737 group : list, numpy 1-D array, pandas Series or None, optional (default=None)
738 Group/query size for Dataset.
739 init_score : list, numpy 1-D array, pandas Series or None, optional (default=None)
740 Init score for Dataset.
741 silent : bool, optional (default=False)
742 Whether to print messages during construction.
743 feature_name : list of strings or 'auto', optional (default="auto")
744 Feature names.
745 If 'auto' and data is pandas DataFrame, data columns names are used.
746 categorical_feature : list of strings or int, or 'auto', optional (default="auto")
747 Categorical features.
748 If list of int, interpreted as indices.
749 If list of strings, interpreted as feature names (need to specify ``feature_name`` as well).
750 If 'auto' and data is pandas DataFrame, pandas unordered categorical columns are used.
751 All values in categorical features should be less than int32 max value (2147483647).
752 Large values could be memory consuming. Consider using consecutive integers starting from zero.
753 All negative values in categorical features will be treated as missing values.
754 The output cannot be monotonically constrained with respect to a categorical feature.
755 params : dict or None, optional (default=None)
756 Other parameters for Dataset.
757 free_raw_data : bool, optional (default=True)
758 If True, raw data is freed after constructing inner Dataset.
759 """
760 self.handle = None
761 self.data = data
762 self.label = label
763 self.reference = reference
764 self.weight = weight
765 self.group = group
766 self.init_score = init_score
767 self.silent = silent
768 self.feature_name = feature_name
769 self.categorical_feature = categorical_feature
770 self.params = copy.deepcopy(params)
771 self.free_raw_data = free_raw_data
772 self.used_indices = None
773 self.need_slice = True
774 self._predictor = None

Callers 4

_construct_datasetMethod · 0.70
create_validMethod · 0.70
subsetMethod · 0.70
refitMethod · 0.70

Calls

no outgoing calls

Tested by

no test coverage detected