Method expression

qlib/data/data.py:843–879

(self, instrument, field, start_time=None, end_time=None, freq="day")

        self.time2idx = time2idx

    def expression(self, instrument, field, start_time=None, end_time=None, freq="day"):
        expression = self.get_expression_instance(field)
        start_time = time_to_slc_point(start_time)
        end_time = time_to_slc_point(end_time)

        # Two kinds of queries are supported
        # - Index-based expression: this may save a lot of memory because the datetime index is not saved on the disk
        # - Data with datetime index expression: this will make it more convenient to integrating with some existing databases
        if self.time2idx:
            _, _, start_index, end_index = Cal.locate_index(start_time, end_time, freq=freq, future=False)
            lft_etd, rght_etd = expression.get_extended_window_size()
            query_start, query_end = max(0, start_index - lft_etd), end_index + rght_etd
        else:
            start_index, end_index = query_start, query_end = start_time, end_time

        try:
            series = expression.load(instrument, query_start, query_end, freq)
        except Exception as e:
            get_module_logger("data").debug(
                f"Loading expression error: "
                f"instrument={instrument}, field=({field}), start_time={start_time}, end_time={end_time}, freq={freq}. "
                f"error info: {str(e)}"
            )
            raise
        # Ensure that each column type is consistent
        # FIXME:
        # 1) The stock data is currently float. If there is other types of data, this part needs to be re-implemented.
        # 2) The precision should be configurable
        try:
            series = series.astype(np.float32)
        except ValueError:
            pass
        except TypeError:
            pass
        if not series.empty:
            series = series.loc[start_index:end_index]
        return series


class LocalDatasetProvider(DatasetProvider):