LSTM using pandas 0.25 and tensorflow 1.15

General Discussion
#1

I have created this question in detail on stackoverflow, but as a synopsis:

I want to create an LSTM which will take in a pandas dataframe of independent variables and optimize for the dependent variable. The error I’m receiving is as such:

KeyError: ‘No key found for either mapped or original key. Mapped Key: []; Original Key: []’

Unfortunately I cannot update tensorflow from 1.15 or pandas from 0.25 as the environment I’m developing in doesn’t allow for it. The error occurs in the fit_generator() method.

#2

A post was merged into an existing topic: LSTM using pandas 0.25 and tensorflow 1.15

#3