Skip to content

Commit 56af60d

Browse files
ahn1340mfeurer
authored andcommitted
Circle Drop (#575)
* Extending Autosklearn. First commit. * Add regression example * CI: upper bound numpy version due to travis failures * CI: upper bound numpy version due to travis failures * use tempfile.gettempdir() (#521) * use tempfile.gettempdir() * follow quality review coding standards * Remove a colon from README.md (#527) * fixing warnings on non-tuple sequence for indexing (#526) * fix string formatting (#540) * FIX removing models wrt wrong metric in ensemble (#522) * Add examples for extending auto-sklearn. * . * Change datasets used in examples from digits to breast_cancer. * First commit * Fixing codacy errors * Fixing bug * [Debug] try different numpy version * [Debug] Try with latest numpy version * Set numpy version to 1.14.5 * First commit * Fixing bug * Modify flake8_diff.sh * Extending Autosklearn. First commit. * Add regression example * Add examples for extending auto-sklearn. * . * Fixing codacy errors * Change example (#553) * Change datasets used in examples from digits to breast_cancer. * [Debug] try different numpy version * [Debug] Try with latest numpy version * Set numpy version to 1.14.5 * Fix line length in exanple_parallel.py * [WIP]Add argument for custom logger configuration. (#505) * Add argument for custom logger configuration. First commit, work in progress. * Minor changes. * Modify suggested changes [WIP] * . * . * . * . * Fix minor details * Fix travis not recognizing example_config.yaml * . * . * . * . * . * Change datasets used in examples from digits to breast_cancer. * Fix codacy error * Revert codacy error fixing * [Debug] check if numpy causes error * [Debug] experimenting with numpy * [Debug] try to manually install libgcc * [Debug] libgcc * [Debug] libgcc * [Debug] libgcc * [Debug] libgcc * [Debug]. * [Debug]. * [Debug] . * [Debug]. * [Debug]. * [Debug] used older numpy version * [Debug] numpy * [Debug] try numpy version 1.14.6 * [Debug] try different numpy version * [Debug] Try with latest numpy version * Set numpy version to 1.14.5 * Add argument for custom logger configuration. First commit, work in progress. * Minor changes. * Modify suggested changes [WIP] * . * . * . * . * Fix minor details * Fix travis not recognizing example_config.yaml * . * . * . * . * . * Fix codacy error * Revert codacy error fixing * [Debug] experimenting with numpy * [Debug] try to manually install libgcc * [Debug] libgcc * [Debug] libgcc * [Debug] libgcc * [Debug] libgcc * [Debug]. * [Debug]. * [Debug] . * [Debug]. * [Debug]. * [Debug] used older numpy version * [Debug] try numpy version 1.14.6 * Delete libgcc_check.sh used for debugging. * Fix numpy version and remove blank lines * Fix line length in example_parallel.py * Fix minor error * FIX #566: sort ensemble correctly (#567) * Fix Line length in example_parallel.py * Fix line length in example_parallel.py * Fix minor error * Fix codacy error "parameters differ from overriden 'fit' method" * Check target type at the beginning of the fitting process. (#506) * Check target type at the beginning of the fitting process. * . * Fixed minor error in uniitest * . * Add unittest for target type checking. * . * . * Change datasets used in examples from digits to breast_cancer. * [Debug] try with numpy version 1.14.5 * [Debug] Check if numpy version 1.14.6 raises error. * [Debug] try different numpy version * [Debug] Try with latest numpy version * Set numpy version to 1.14.5 * Check target type at the beginning of the fitting process. * . * Fixed minor error in uniitest * . * Add unittest for target type checking. * . * . * [Debug] Check if numpy version 1.14.6 raises error. * Fix numpy version to 1.14.5 * Add comment to Mock in test_type_of_target * Fix line length in example_parallel.py * Fix minor error * Update test_automl.py * Add python 3.7to Travis, change python_requirement in setup.py. * Add solver hyperparameter in MLP classifier example, increase runtime of classifier in example_extend_preprocessing.py * Change all occurences of master to development in flake8_diff.sh * numpy requirement is now >=1.9.0<=1.14.5 * Fix requirement inequality mistake * change initial numpy version to 1.14.5. * Deploy using travis instead of circle * FIX error in travis.yml caused by stashing * Test that deploy works. * Debugging. Set local_dir to doc/development. * Done Testing. Finalize the PR. * Delete circle_install.sh
1 parent c963f75 commit 56af60d

File tree

5 files changed

+73
-126
lines changed

5 files changed

+73
-126
lines changed

.travis.yml

Lines changed: 12 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -16,7 +16,7 @@ matrix:
1616
- os: linux
1717
env: DISTRIB="conda" PYTHON_VERSION="3.5" COVERAGE="true" MINICONDA_URL="https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh"
1818
- os: linux
19-
env: DISTRIB="conda" PYTHON_VERSION="3.6" MINICONDA_URL="https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh"
19+
env: DISTRIB="conda" PYTHON_VERSION="3.6" DOCPUSH="true" MINICONDA_URL="https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh"
2020
- os: linux
2121
env: DISTRIB="conda" PYTHON_VERSION="3.6" EXAMPLES="true" MINICONDA_URL="https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh"
2222
- os: linux
@@ -77,5 +77,15 @@ install:
7777
- python setup.py install
7878

7979
script: bash ci_scripts/test.sh
80-
after_success: source ci_scripts/success.sh
80+
after_success: source ci_scripts/success.sh && source ci_scripts/create_doc.sh $TRAVIS_BRANCH "doc_result"
8181

82+
deploy:
83+
provider: pages
84+
skip-cleanup: true
85+
github-token: $GITHUB_TOKEN # set in the settings page of my repository
86+
keep-hisotry: true
87+
commiter-from-gh: true
88+
on:
89+
all_branches: true
90+
condition: $doc_result = "success"
91+
local_dir: doc/$TRAVIS_BRANCH

ci_scripts/circle_install.sh

Lines changed: 0 additions & 20 deletions
This file was deleted.

ci_scripts/create_doc.sh

Lines changed: 61 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,61 @@
1+
# This script is mostly adopted from https://github.com/openml/openml-python/blob/develop/ci_scripts/create_doc.sh
2+
3+
set -euo pipefail
4+
5+
# Check if DOCPUSH is set
6+
if ! [[ -z ${DOCPUSH+x} ]]; then
7+
8+
if [[ "$DOCPUSH" == "true" ]]; then
9+
10+
# install documentation building dependencies
11+
pip install --upgrade matplotlib seaborn setuptools nose coverage sphinx pillow sphinx-gallery sphinx_bootstrap_theme cython numpydoc nbformat nbconvert mock
12+
13+
# $1 is the branch name
14+
# $2 is the global variable where we set the script status
15+
16+
if ! { [ $1 = "master" ] || [ $1 = "development" ]; }; then
17+
{ echo "Not one of the allowed branches"; exit 0; }
18+
fi
19+
20+
# delete any previous documentation folder
21+
if [ -d doc/$1 ]; then
22+
rm -rf doc/$1
23+
fi
24+
25+
# create the documentation
26+
cd doc && make html 2>&1
27+
28+
# create directory with branch name
29+
# the documentation for dev/stable from git will be stored here
30+
mkdir $1
31+
32+
# get previous documentation from github
33+
git clone https://github.com/automl/auto-sklearn.git --branch gh-pages --single-branch
34+
35+
# copy previous documentation
36+
cp -r auto-sklearn/. $1
37+
rm -rf auto-sklearn
38+
39+
# if the documentation for the branch exists, remove it
40+
if [ -d $1/$1 ]; then
41+
rm -rf $1/$1
42+
fi
43+
44+
# copy the updated documentation for this branch
45+
mkdir $1/$1
46+
cp -r build/html/. $1/$1
47+
48+
# takes a variable name as an argument and assigns the script outcome to a
49+
# variable with the given name. If it got this far, the script was successful
50+
function set_return() {
51+
# $1 is the variable where we save the script outcome
52+
local __result=$1
53+
local status='success'
54+
eval $__result="'$status'"
55+
}
56+
57+
set_return "$2"
58+
fi
59+
fi
60+
# Workaround for travis failure
61+
set +u

ci_scripts/push_doc.sh

Lines changed: 0 additions & 42 deletions
This file was deleted.

circle.yml

Lines changed: 0 additions & 62 deletions
This file was deleted.

0 commit comments

Comments
 (0)