Get the raw data of the Dataset. Returns ------- data : string, numpy array, pandas DataFrame, H2O DataTable's Frame, scipy.sparse, list of numpy arrays or None Raw data used in the Dataset construction.
(self)
| 1516 | return self.init_score |
| 1517 | |
| 1518 | def get_data(self): |
| 1519 | """Get the raw data of the Dataset. |
| 1520 | |
| 1521 | Returns |
| 1522 | ------- |
| 1523 | data : string, numpy array, pandas DataFrame, H2O DataTable's Frame, scipy.sparse, list of numpy arrays or None |
| 1524 | Raw data used in the Dataset construction. |
| 1525 | """ |
| 1526 | if self.handle is None: |
| 1527 | raise Exception("Cannot get data before construct Dataset") |
| 1528 | if self.need_slice and self.used_indices is not None and self.reference is not None: |
| 1529 | self.data = self.reference.data |
| 1530 | if self.data is not None: |
| 1531 | if isinstance(self.data, np.ndarray) or scipy.sparse.issparse(self.data): |
| 1532 | self.data = self.data[self.used_indices, :] |
| 1533 | elif isinstance(self.data, DataFrame): |
| 1534 | self.data = self.data.iloc[self.used_indices].copy() |
| 1535 | elif isinstance(self.data, DataTable): |
| 1536 | self.data = self.data[self.used_indices, :] |
| 1537 | else: |
| 1538 | warnings.warn("Cannot subset {} type of raw data.\n" |
| 1539 | "Returning original raw data".format(type(self.data).__name__)) |
| 1540 | self.need_slice = False |
| 1541 | if self.data is None: |
| 1542 | raise LightGBMError("Cannot call `get_data` after freed raw data, " |
| 1543 | "set free_raw_data=False when construct Dataset to avoid this.") |
| 1544 | return self.data |
| 1545 | |
| 1546 | def get_group(self): |
| 1547 | """Get the group of the Dataset. |