1 import pandas as pd
2 import datetime as datetime
----> 3 from flood_forecast.trainer import train_function
4 import wandb
5 import os
...
from numba.np.ufunc import _internal
from numba.np.ufunc.parallel import ParallelUFuncBuilder, ParallelGUFuncBuilder
from numba.core.registry import DelayedRegistry
SystemError: initialization of _internal failed without raising an exception
Tried several configurations of numpy (1.23.5 and 1.23.0 and 1.22.5) and numba (0.56.4) due to following issue reported on numba https://github.com/numba/numba/issues/8718 which did not solve the problem. Thinking an update to the requirements.txt might fix the issue.
Error message when importing train function
SystemError Traceback (most recent call last)
----> 3 from flood_forecast.trainer import train_function 4 import wandb 5 import os
... from numba.np.ufunc import _internal from numba.np.ufunc.parallel import ParallelUFuncBuilder, ParallelGUFuncBuilder from numba.core.registry import DelayedRegistry
SystemError: initialization of _internal failed without raising an exception
Tried several configurations of numpy (1.23.5 and 1.23.0 and 1.22.5) and numba (0.56.4) due to following issue reported on numba https://github.com/numba/numba/issues/8718 which did not solve the problem. Thinking an update to the requirements.txt might fix the issue.
Using python 3.8.15