Github user srowen commented on a diff in the pull request:
https://github.com/apache/spark/pull/13516#discussion_r65818408
--- Diff: docs/mllib-linear-methods.md ---
@@ -257,10 +257,10 @@ applying the logistic function
\mathrm{f}(z) = \frac{1}{1 + e^{-z}}
\]`
where $z = \wv^T \x$.
-By default, if $\mathrm{f}(\wv^T x) > 0.5$, the outcome is positive, or
-negative otherwise, though unlike linear SVMs, the raw output of the
logistic regression
-model, $\mathrm{f}(z)$, has a probabilistic interpretation (i.e., the
probability
-that $\x$ is positive).
+By default, if $\mathrm{f}(\wv^T x) > 0.5$, the outcome is positive, else
it is negative.
+Logistic regression is distinct from say linear SVMs in its formally being
a Bayesian model, albeit trivial one: rather than producing directly an
'input-output machine', the conditional distribution of the output given the
input is modeled explicitly through the function $\mathrm{f}$ above; this model
can be and is then used to provide definite outputs for definite inputs.
--- End diff --
Even if this is accurate, I don't think it's an improvement. This drops the
key points, that the output of logistic regression may be interpreted as a
probability. I don't believe this description clarifies anything for Spark
users.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
