I was wondering if it is possible to use the flow_from_dataframe on Keras with multiple inputs. I have a Dataframe with 3 features – TIF image names, and 10 labels – values, and I want to create a CNN with regression. I’ve read the documentation and I created my test and validation generators, with y_col as my multiple label columns. However, I do not understand how I can use multiple x columns.

I tried to pass a list of columns e.g.:

x_col = [feature1 , feature2 , feature3]
y_col = [label1, label2,…,label10]

but I have an error: All values in column x_col must be strings error.

I’d really appreciate it if someone could help me to understand how to accomplish that. Thanks in advance.

  • $\begingroup$ The error comes as it's only expecting a single string. (x_col: string, column in data-frame that contains the filenames); Why don't you write your own variant, it's not that difficult, right? $\endgroup$ – Aditya Jul 26 at 14:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.