TypeError: int() argument must be a string, a bytes-like object or a number, not 'NaTType'

Data
#1

Hey Im new to using zipline. I am trying to run one of the examples and getting an error when importing using jupyter notebook via install in conda.

Is this module dated or am I doing something wrong? Thanks!

from zipline.api import order_target, record, symbol

TypeError Traceback (most recent call last)
Input In [3], in <cell line: 1>()
----> 1 from zipline.api import order_target, record, symbol

File C:\ProgramData\Anaconda3\lib\site-packages\zipline_init_.py:21, in
17 import numpy as np
19 # This is not a place to dump arbitrary classes/modules for convenience,
20 # it is a place to expose the public interfaces.
—> 21 from trading_calendars import get_calendar
23 from . import data
24 from . import finance

File C:\ProgramData\Anaconda3\lib\site-packages\trading_calendars_init_.py:16, in
1 #
2 # Copyright 2018 Quantopian, Inc.
3 #
(…)
13 # See the License for the specific language governing permissions and
14 # limitations under the License.
—> 16 from .trading_calendar import TradingCalendar
17 from .calendar_utils import (
18 clear_calendars,
19 deregister_calendar,
(…)
25 resolve_alias,
26 )
28 all = [
29 ‘clear_calendars’,
30 ‘deregister_calendar’,
(…)
37 ‘TradingCalendar’,
38 ]

File C:\ProgramData\Anaconda3\lib\site-packages\trading_calendars\trading_calendar.py:33, in
30 from pandas.tseries.offsets import CustomBusinessDay
31 import toolz
—> 33 from .calendar_helpers import (
34 NP_NAT,
35 compute_all_minutes,
36 next_divider_idx,
37 previous_divider_idx,
38 )
39 from .utils.memoize import lazyval
40 from .utils.pandas_utils import days_at_time

File C:\ProgramData\Anaconda3\lib\site-packages\trading_calendars\calendar_helpers.py:6, in
2 import pandas as pd
4 NANOSECONDS_PER_MINUTE = int(6e10)
----> 6 NP_NAT = np.array([pd.NaT], dtype=np.int64)[0]
9 def next_divider_idx(dividers, minute_val):
11 divider_idx = np.searchsorted(dividers, minute_val, side=“right”)

TypeError: int() argument must be a string, a bytes-like object or a number, not ‘NaTType’

#2

in the conda CMD if I simply enter the command ‘zipline’ I get an identical error:

error
error924×219 6.94 KB

I just updated to conda 4.13.0 tonight, maybe thats why its throwing the error?

#3

Had the same issue…googled it, and the following hack worked, yet not really sure why…so use with discretion!

In file:
/miniconda3/envs/zrl/lib/python3.8/site-packages/trading_calendars/calendar_helpers.py

Change:
#new NP_NAT = np.array([pd.NaT])
# orig NP_NAT = np.array([pd.NaT], dtype=np.int64)[0]
# ajjc old issue #40 2022-08-09
NP_NAT = np.array([pd.NaT.asm8.view(‘i8’)], dtype=np.int64)[0]

alan