aigamedev / scikit-neuralnetwork
100%
master: 100%

DEFAULT BRANCH: master

LAST BUILD BRANCH: lasagne

Repo Added 01 Apr 2015 12:11PM UTC

Total Files 20

# Builds 492 Last

Badge

Embed ▾

LAST BUILD ON BRANCH conv
branch: conv

CHANGE BRANCH
x

Reset

conv

master

pn

adagrad

serialization

params

power

perf

sparse

init

wrap

theano

0.1

v0.1

profile

regularize

loss

autoencoder

nan

testing

sklearn

RNN

pretrain

logging

digits

v0.2

training

deepy

backend

fix_83

multi-label

mutator

color

v0.3

lasagne

v0.4

callback

cost

outconv

ELU

v0.5

threads

voting

monkeypatch

batch-size

v0.6

v0.6.1

memmap

outshp

pandas

batchnorm

native

v0.7

Committed 4 Jun 2015 - 15:27 coverage remained the same at 100.0%

Build # 369

Build Type

push

travis-ci

Committed by

alexjc

Commit Message

Passing in batch_size to the MLP, which helps creating convolution layers that require a batch size: eg. border_mode='full'.  There are still many issues getting pooling to work, since it requires good knowledge of the size of each layer and output.

Updates #68.

Run Details

1395 of 1395 relevant lines covered (100.0%)

2.0 hits per line

Relevant lines Covered

1395 RELEVANT LINES 1395 COVERED LINES

2.0 HITS PER LINE

Source Files on conv

Coverage	∆	File	Lines	Relevant	Covered	Missed	Hits/Line

Recent Builds

Builds	Branch	Coverage	Commit	Committer	Type	Time	Via
369	conv	100.0	Passing in batch_size to the MLP, which helps creating convolution layers that require a batch size: eg. border_mode='full'. There are still many issues getting pooling to work, since it requires good knowledge of the size of each layer and outpu...	alexjc	push	04 Jun 2015 03:36PM UTC	travis-ci
362	conv	100.0	Now correctly serialising a convolution layers, and tests for serialisation determinism on CNN. Fixes #67.	alexjc	push	04 Jun 2015 02:22PM UTC	travis-ci
247	conv	100.0	Fix for the Python3 build, logging tuples for shape correctly.	alexjc	push	06 May 2015 12:09PM UTC	travis-ci
245	conv	100.0	Fixed convolution tests and pool initialization defaults.	alexjc	push	06 May 2015 09:11AM UTC	travis-ci
243	conv	98.99	Further improvements for convolution, in particular disabling pooling if not specified explicitly. Also using some asserts as sanity checks, though unit_count calculation is <=1 off when pooling is used. This is only used for weight initialization...	alexjc	push	06 May 2015 08:59AM UTC	travis-ci

See All Builds (490)

Badge your Repo: scikit-neuralnetwork

We detected this repo isn’t badged! Grab the embed code to the right, add it to your repo to show off your code coverage, and when the badge is live hit the refresh button to remove this message.

Embed ▾

Refresh

aigamedev / scikit-neuralnetwork
100%
master: 100%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

CHANGE BRANCH
x

Relevant lines Covered

Source Files on conv

Recent Builds

Badge your Repo: scikit-neuralnetwork

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

aigamedev / scikit-neuralnetwork 100% master: 100%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

CHANGE BRANCH x

Relevant lines Covered

Source Files on conv

Recent Builds

Badge your Repo: scikit-neuralnetwork

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

aigamedev / scikit-neuralnetwork
100%
master: 100%

README BADGES
x

CHANGE BRANCH
x

README BADGES
x