mirror of
https://github.com/saymrwulf/zipline.git
synced 2026-05-16 21:10:11 +00:00
e491ebef53
This commit adds initial support for international markets in the Pipeline API
via addition of the concept of "domains".
Significant changes on this branch include:
- Added `zipline.pipeline.domain` and the `Domain` interface.
- Added several canonical domain definitions:
- GENERIC
- US_EQUITIES
- CA_EQUITIES
- GB_EQUITIES
- Added `zipline.country` as a place to define canonical country code names.
- Made `PipelineLoader` an interface and added new `domain` parameter to
signature of `load_adjusted_array`.
- Added a new system for "generic" and "specialized" DataSet and BoundColumn
objects.
- Added a new system for inferring a domain from a pipeline that contains a mix
of generic and non-generic terms.
- Reworked the built-in pricing dataset. USEquityPricing dataset is now a
specialized version of a generic EquityPricing dataset, and most built-in
factors are now implemented in terms of EquityPricing rather than
USEquityPricing.
- Removed `zipline.data.us_equity_pricing`.
- Moved BcolzDailyBar{Reader,Writer} to `zipline.data.bcolz_daily_bars`.
- Moved SQLiteAdjustment{Reader,Writer} to `zipline.data.adjustments`.
New dependencies added as part of this work:
- iso3166 (for canonical country code definitions)
- python-interface (for strict interface definitions).
See https://github.com/quantopian/zipline/issues/2265 for a full description of
the design for domains.
50 lines
1.4 KiB
Python
50 lines
1.4 KiB
Python
from numpy import (
|
|
float64,
|
|
uint32,
|
|
int64,
|
|
)
|
|
from bcolz import ctable
|
|
|
|
from zipline.data.bcolz_daily_bars import (
|
|
BcolzDailyBarWriter,
|
|
OHLC,
|
|
UINT32_MAX
|
|
)
|
|
|
|
|
|
class DailyBarWriterFromDataFrames(BcolzDailyBarWriter):
|
|
_csv_dtypes = {
|
|
'open': float64,
|
|
'high': float64,
|
|
'low': float64,
|
|
'close': float64,
|
|
'volume': float64,
|
|
}
|
|
|
|
def __init__(self, asset_map):
|
|
self._asset_map = asset_map
|
|
|
|
def gen_tables(self, assets):
|
|
for asset in assets:
|
|
yield asset, ctable.fromdataframe(assets[asset])
|
|
|
|
def to_uint32(self, array, colname):
|
|
arrmax = array.max()
|
|
if colname in OHLC:
|
|
self.check_uint_safe(arrmax * 1000, colname)
|
|
return (array * 1000).astype(uint32)
|
|
elif colname == 'volume':
|
|
self.check_uint_safe(arrmax, colname)
|
|
return array.astype(uint32)
|
|
elif colname == 'day':
|
|
nanos_per_second = (1000 * 1000 * 1000)
|
|
self.check_uint_safe(arrmax.view(int64) / nanos_per_second,
|
|
colname)
|
|
return (array.view(int64) / nanos_per_second).astype(uint32)
|
|
|
|
@staticmethod
|
|
def check_uint_safe(value, colname):
|
|
if value >= UINT32_MAX:
|
|
raise ValueError(
|
|
"Value %s from column '%s' is too large" % (value, colname)
|
|
)
|