I’ve been working through part 1 of the course and I had a question about a dataset I’ve come across. The dataset is tabular data and one of the columns is a historgram array. The lengths of the array vary from row to row. What is the best way to handle this? I’ve thought of a couple different things:
- normalize (sum the values in the array to 1 singular value, or get an average)
- Find the max length array and append 0’s to every other array to match that length
Is this a required step or can DL models handle variable length arrays as data types ?
Thanks for the help and clarification