You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Potentially similar to issue #88 but seems like a distinct issue.
The __init__ method for the Network class creates a node_idx attribute as a pd.Series whose dtype is specified as "int". This yields a Series with int64 as its dtype. I don't know if this is due to a version update in pandas or numpy or something else, but it breaks the call to cyaccess, which expects node_idx to be long but is now getting long long. I tried coercing all inputs to np.int32, but it did not resolve the issue. The problem seems to be the dtype="int" in line 86 of network.py. If I change it to dtype=np.int32 my code executes. This seems like a more explicit specification in any case, so it may be preferable to simply having "int".
The text was updated successfully, but these errors were encountered:
Potentially similar to issue #88 but seems like a distinct issue.
The
__init__
method for theNetwork
class creates anode_idx
attribute as apd.Series
whosedtype
is specified as"int"
. This yields a Series withint64
as its dtype. I don't know if this is due to a version update in pandas or numpy or something else, but it breaks the call tocyaccess
, which expectsnode_idx
to belong
but is now gettinglong long
. I tried coercing all inputs tonp.int32
, but it did not resolve the issue. The problem seems to be thedtype="int"
in line 86 of network.py. If I change it todtype=np.int32
my code executes. This seems like a more explicit specification in any case, so it may be preferable to simply having"int"
.The text was updated successfully, but these errors were encountered: