use softmax eq in predict_probability func

[tomofumi-nakano]

No description provided.

[cjlin1]

Thanks. However, please note that in liblinear we didn't
do max entropy. For multiclass we do 1vs rest and need to couple
prob outputs of k binary classifiers. The correct way should be
like
sec 4.1 of
http://www.csie.ntu.edu.tw/~cjlin/papers/generalBT.pdf
but we somehow took a simplified version of that

中野智文 writes:

---

You can view, comment on, or merge this pull request online at:

#20

Commit Summary

• use softmax eq in predict_probability func

File Changes

• M linear.cpp (5)

Patch Links:

• https://github.com/cjlin1/liblinear/pull/20.patch
• https://github.com/cjlin1/liblinear/pull/20.diff

—
You are receiving this because you are subscribed to this
thread.
Reply to this email directly or view it on GitHub*