2
97%
master: 97%

Ran 21 Dec 2016 04:57PM UTC

Files 13

Run time 1s

Badge

Committed 21 Dec 2016 04:55PM UTC coverage: 97.425% (-0.9%) from 98.303%

Job # DISTRIB="conda" PYTHON_VERSION="3.5" COVERAGE="true" NUMPY_VERSION="1.10.4" SCIPY_VERSION="0.17.0" CYTHON_VERSION="0.23.4" SKLEARN_VERSION="0.17.1"

Build Type

Pull #8

travis-ci

Committed by

web-flow

Commit Message

Update scaling and initialization.

Regularization scaling is now ON by default. I think this is
sensible, because it keeps the choice independent of data split.

Adagrad seems very sensitive to the initial norm of P, so I changed
the init to have unit variance rather than 0.01.
Makes benchmark more reasonable but norms are still weird.
Finnicky tests (fm warm starts) had to be updated, but most
things behave well.

Pull Request Pull Request #8: [WIP] Adagrad solver for arbitrary orders.

Run Details

946 of 971 relevant lines covered (97.43%)

0.97 hits per line

Source Files on job 82.2 (DISTRIB="conda" PYTHON_VERSION="3.5" COVERAGE="true" NUMPY_VERSION="1.10.4" SCIPY_VERSION="0.17.0" CYTHON_VERSION="0.23.4" SKLEARN_VERSION="0.17.1")

scikit-learn-contrib / polylearn / 82 / 2
97%
master: 97%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 82.2 (DISTRIB="conda" PYTHON_VERSION="3.5" COVERAGE="true" NUMPY_VERSION="1.10.4" SCIPY_VERSION="0.17.0" CYTHON_VERSION="0.23.4" SKLEARN_VERSION="0.17.1")

scikit-learn-contrib / polylearn / 82 / 2 97% master: 97%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 82.2 (DISTRIB="conda" PYTHON_VERSION="3.5" COVERAGE="true" NUMPY_VERSION="1.10.4" SCIPY_VERSION="0.17.0" CYTHON_VERSION="0.23.4" SKLEARN_VERSION="0.17.1")

scikit-learn-contrib / polylearn / 82 / 2
97%
master: 97%

README BADGES
x