Github user njayaram2 commented on a diff in the pull request:
https://github.com/apache/madlib/pull/250#discussion_r178168247
--- Diff: src/ports/postgres/modules/convex/mlp_igd.py_in ---
@@ -667,7 +678,8 @@ def _validate_dependent_var(source_table,
dependent_varname,
if is_classification:
# Currently, classification doesn't accept an
# array for dep type in IGD
- _assert("[]" not in expr_type and expr_type in
classification_types,
+ _assert(("[]" in expr_type and expr_type[:-2] in int_types) \
--- End diff --
- One-hot encoding is normally supposed to be an integer array, hence was
checking only for int type. But, that assumption might be too strict, so will
make changes to allow for other numeric types too.
- Good catch. It might be a good idea to disallow anything greater than a
1-D array. Will make an error check for the same.
---
