Github user mktal commented on a diff in the pull request:
https://github.com/apache/incubator-madlib/pull/48#discussion_r69011515
--- Diff: src/ports/postgres/modules/svm/svm.py_in ---
@@ -676,27 +985,32 @@ def svm_help(schema_madlib, message, is_svc,
**kwargs):
def svm(schema_madlib, source_table, model_table,
dependent_varname, independent_varname, kernel_func,
kernel_params, grouping_col, params, is_svc,
- verbose, detect_novelty=False, **kwargs):
+ verbose, add_intercept=False, **kwargs):
--- End diff --
One thing still not very clear to me. `add_intercept` being true override
`fit_intercept` which is intended. But `add_intercept` being false will have no
effect on `fit_intercept`. Also if user uses `array[1, ...]` and linear kernel
and specify `fit_intercept=true` in kernel params, the algorithm will
essentially fit `array[1,1,...]`? I thought the behavior in the linear case is
such that `fit_intercept` in the kernel params is a `no-op` while in the kernel
case `fit_intercept` overrides. There are some inconsistencies due to multiple
ways of specifying whether to fit intercept. I am a bit confused and need to
think about it more....
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
