Getting a TypeError: '<' not supported between instances of 'float' and 'complex' training SNARIMAX #947
Unanswered
RaphaelChevallier
asked this question in
Q&A
Replies: 1 comment 2 replies
-
Hey there. First of all you can't pass an |
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello,
I am trying to implement a snarimax time series model. I have regular spaced intervals of data. I have exactly 179 dicts points of data every 15 mins. I use a for loop to train the model one dict at a time then forecast the next 15 mins value.
However during this for loop training model I get some dicts that issue an error on the training method model.learn_one.
TypeError: '<' not supported between instances of 'float' and 'complex'
. This is odd as they are all the same dict values. Every entry has the same key values and the value types are always floats. I have tried to handle the error and it is very random as to what data throws thisAny ideas would be much appreciated
Beta Was this translation helpful? Give feedback.
All reactions