I’m trying to train two identical models on two almost identical data sets. The problem is though the training starts out completely different and has a completely different outcome. My two datasets have the exact number of items and only differ slightly in some values since they are calculated at two different sources, but the differences are only on a dozen or so of about ~100 columns and differ by either more decimal places or by a very small amount. e.g. 1566.50125 vs 1566.5 and 1723.9 vs 1714.75.
Are there any good methods for finding all of the differences between two dataframes? Order, types, and data differences. I’m thinking there is something else I’m overlooking between the two.
Thanks