97 lines
3.1 KiB
Python
97 lines
3.1 KiB
Python
# piker: trading gear for hackers
|
|
# Copyright (C) 2018-present Tyler Goodlet (in stewardship for pikers)
|
|
|
|
# This program is free software: you can redistribute it and/or modify
|
|
# it under the terms of the GNU Affero General Public License as published by
|
|
# the Free Software Foundation, either version 3 of the License, or
|
|
# (at your option) any later version.
|
|
|
|
# This program is distributed in the hope that it will be useful,
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
# GNU Affero General Public License for more details.
|
|
|
|
# You should have received a copy of the GNU Affero General Public License
|
|
# along with this program. If not, see <https://www.gnu.org/licenses/>.
|
|
|
|
"""
|
|
numpy data source coversion helpers.
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
from bidict import bidict
|
|
import numpy as np
|
|
|
|
|
|
def_iohlcv_fields: list[tuple[str, type]] = [
|
|
|
|
# YES WE KNOW, this isn't needed in polars but we use it for doing
|
|
# ring-buffer like pre/append ops our our `ShmArray` real-time
|
|
# numpy-array buffering system such that there is a master index
|
|
# that can be used for index-arithmetic when write data to the
|
|
# "middle" of the array. See the ``tractor.ipc.shm`` pkg for more
|
|
# details.
|
|
('index', int),
|
|
|
|
# presume int for epoch stamps since it's most common
|
|
# and makes the most sense to avoid float rounding issues.
|
|
# TODO: if we want higher reso we should use the new
|
|
# ``time.time_ns()`` in python 3.10+
|
|
('time', int),
|
|
('open', float),
|
|
('high', float),
|
|
('low', float),
|
|
('close', float),
|
|
('volume', float),
|
|
|
|
# TODO: can we elim this from default field set to save on mem?
|
|
# i think only kraken really uses this in terms of what we get from
|
|
# their ohlc history API?
|
|
# ('bar_wap', float), # shouldn't be default right?
|
|
]
|
|
|
|
# remove index field
|
|
def_ohlcv_fields: list[tuple[str, type]] = def_iohlcv_fields.copy()
|
|
def_ohlcv_fields.pop(0)
|
|
assert (len(def_iohlcv_fields) - len(def_ohlcv_fields)) == 1
|
|
|
|
# TODO: for now need to construct this manually for readonly arrays, see
|
|
# https://github.com/numba/numba/issues/4511
|
|
# from numba import from_dtype
|
|
# base_ohlc_dtype = np.dtype(def_ohlc_fields)
|
|
# numba_ohlc_dtype = from_dtype(base_ohlc_dtype)
|
|
|
|
# map time frame "keys" to seconds values
|
|
tf_in_1s = bidict({
|
|
1: '1s',
|
|
60: '1m',
|
|
60*5: '5m',
|
|
60*15: '15m',
|
|
60*30: '30m',
|
|
60*60: '1h',
|
|
60*60*24: '1d',
|
|
})
|
|
|
|
|
|
def ohlc_zeros(length: int) -> np.ndarray:
|
|
"""Construct an OHLC field formatted structarray.
|
|
|
|
For "why a structarray" see here: https://stackoverflow.com/a/52443038
|
|
Bottom line, they're faster then ``np.recarray``.
|
|
|
|
"""
|
|
return np.zeros(length, dtype=base_ohlc_dtype)
|
|
|
|
|
|
def _nan_to_closest_num(array: np.ndarray):
|
|
"""Return interpolated values instead of NaN.
|
|
|
|
"""
|
|
for col in ['open', 'high', 'low', 'close']:
|
|
mask = np.isnan(array[col])
|
|
if not mask.size:
|
|
continue
|
|
array[col][mask] = np.interp(
|
|
np.flatnonzero(mask), np.flatnonzero(~mask), array[col][~mask]
|
|
)
|