Update README, travis files, and tests.py

change test
llmmddCoder · Feb 18, 2016 · df0f07f · df0f07f
1 parent cf579a5
commit df0f07f
Show file tree

Hide file tree

Showing 5 changed files with 18 additions and 10 deletions.
diff --git a/.travis.yml b/.travis.yml
@@ -4,10 +4,10 @@ virtualenv:
 env:
   matrix:
     # let's start simple:
-    - PYTHON_VERSION="2.7" LATEST="true" "DEAP_VERSION=1.0.1"
-    - PYTHON_VERSION="3.4" LATEST="true" "DEAP_VERSION=1.0.1"
-    - PYTHON_VERSION="3.4" COVERAGE="true" LATEST="true" "DEAP_VERSION=1.0.1"
-    - PYTHON_VERSION="3.5" LATEST="true" "DEAP_VERSION=1.0.1"
+    - PYTHON_VERSION="2.7" LATEST="true" "DEAP_VERSION=1.0.1" "XGBOOST_VERSION=0.4a30"
+    - PYTHON_VERSION="3.4" LATEST="true" "DEAP_VERSION=1.0.1" "XGBOOST_VERSION=0.4a30"
+    - PYTHON_VERSION="3.4" COVERAGE="true" LATEST="true" "DEAP_VERSION=1.0.1" "XGBOOST_VERSION=0.4a30"
+    - PYTHON_VERSION="3.5" LATEST="true" "DEAP_VERSION=1.0.1" "XGBOOST_VERSION=0.4a30"
 install: source ./ci/.travis_install.sh
 script: bash ./ci/.travis_test.sh
 after_success:

diff --git a/README.md b/README.md
@@ -43,6 +43,8 @@ TPOT is built on top of several existing Python libraries, including:
 
 * DEAP
 
+* XGBoost
+
 Except for DEAP, all of the necessary Python packages can be installed via the [Anaconda Python distribution](https://www.continuum.io/downloads), which we strongly recommend that you use. We also strongly recommend that you use of Python 3 over Python 2 if you're given the choice.
 
 NumPy, SciPy, pandas, and scikit-learn can be installed in Anaconda via the command:
@@ -51,12 +53,14 @@ NumPy, SciPy, pandas, and scikit-learn can be installed in Anaconda via the comm
 conda install numpy scipy pandas scikit-learn
 ```
 
-DEAP can be installed with `pip` via the command:
+DEAP and XGBoost can be installed with `pip` via the command:
 
 ```Shell
-pip install deap
+pip install deap xgboost
 ```
 
+Optional: For OS X users who want to use OpenMP-enabled compilers to install XGBoost, gcc-5.x.x can be installed with Homebrew: `brew install gcc --without-multilib`.
+
 Finally to install TPOT itself, run the following command:
 
 ```Shell

diff --git a/ci/.travis_install.sh b/ci/.travis_install.sh
@@ -44,6 +44,7 @@ fi
 
 source activate testenv
 pip install deap==$DEAP_VERSION
+pip install xgboost==$XGBOOST_VERSION
 
 if [[ "$COVERAGE" == "true" ]]; then
     pip install coverage coveralls
@@ -56,4 +57,5 @@ python -c "import scipy; print('scipy %s' % scipy.__version__)"
 python -c "import sklearn; print('sklearn %s' % sklearn.__version__)"
 python -c "import pandas; print('pandas %s' % pandas.__version__)"
 python -c "import deap; print('deap %s' % deap.__version__)"
+python -c "import xgboost; print('xgboost %s ' % xgboost.__version__)"
 python setup.py build_ext --inplace
diff --git a/ci/.travis_test.sh b/ci/.travis_test.sh
@@ -15,10 +15,11 @@ python -c "import scipy; print('scipy %s' % scipy.__version__)"
 python -c "import sklearn; print('sklearn %s' % sklearn.__version__)"
 python -c "import pandas; print('pandas %s' % pandas.__version__)"
 python -c "import deap; print('deap %s' % deap.__version__)"
+python -c "import xgboost; print('xgboost %s ' % xgboost.__version__)"
 
 if [[ "$COVERAGE" == "true" ]]; then
-    nosetests -s -v --with-coverage 
+    nosetests -s -v --with-coverage
 else
-    nosetests -s -v 
+    nosetests -s -v
 fi
 #make test-doc test-sphinxext
diff --git a/tests.py b/tests.py
@@ -17,6 +17,7 @@
 from sklearn.linear_model import LogisticRegression
 from sklearn.neighbors import KNeighborsClassifier
 
+from xgboost import XGBClassifier
 
 # Set up the iris data set for testing
 mnist_data = load_digits()
@@ -88,10 +89,10 @@ def test_xgboost():
     '''
 
     tpot_obj = TPOT()
-    result = tpot_obj.xgboost(training_testing_data, 100, 0)
+    result = tpot_obj.xgradient_boosting(training_testing_data, n_estimators=100, learning_rate=0, max_depth=3)
     result = result[result['group'] == 'testing']
 
-    xgb = XGBClassifier(n_estimators=100, max_depth=None, seed=42)
+    xgb = XGBClassifier(n_estimators=100, learning_rate=0.0001, max_depth=3, seed=42)
     xgb.fit(training_features, training_classes)
 
     assert np.array_equal(result['guess'].values, xgb.predict(testing_features))