288
100%

Ran 13 May 2015 05:38PM UTC

Jobs 2

Files 16

Run time 2min

Badge

Embed ▾

pending completion

Build # 288

Build Type

push

travis-ci

Committed by

alexjc

Commit Message

Merge pull request #47 from aigamedev/regularize

Regularization support using L1 and L2, not just dropout.  Can be enabled using the following:

````
from sknn.mlp import Regressor, Layer

nn = Regressor(
    layers=[Layer("Rectifier", units=100), Layer("Linear")],
    regularize='L2',
    weight_decay=0.0001,
    n_iter=10)
nn.fit(X_train, y_train)
````

Each layer also takes a weight_decay `parameter` for customized regularization.  Dropout has now become an additional `regularize` option string.

Closes #44.

Run Details

1092 of 1092 relevant lines covered (100.0%)

2.0 hits per line

Jobs

ID	Job ID	Ran	Files	Coverage
1	288.1	13 May 2015 05:38PM UTC	0	100.0	Travis Job 288.1
2	288.2	13 May 2015 05:40PM UTC	0	100.0	Travis Job 288.2

Source Files on build 288

Detailed source file information is not available for this build.

aigamedev / scikit-neuralnetwork / 288
100%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Jobs

Source Files on build 288

aigamedev / scikit-neuralnetwork / 288 100%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Jobs

Source Files on build 288

aigamedev / scikit-neuralnetwork / 288
100%

README BADGES
x