80 create conda env for chapter 3 fails to build scikit learn 0242

.gitignore

@@ -161,10 +161,12 @@ cython_debug/
 
 #Other - potentially vs code
 *.DS_Store
-**/.DS_Store
+*/.DS_Store
 
 
 #MLFlow etc
 **/artifacts/model
 **/artifacts/** 
-**/mlruns/** 
+**/mlruns/** 
+Chapter08/.DS_Store
+Chapter09/.DS_Store

Chapter03/automl/Dockerfile

@@ -0,0 +1,39 @@
+FROM ubuntu:20.04
+
+# install linux packages
+RUN apt-get update
+
+# Set the locale
+# workaround for https://github.com/automl/auto-sklearn/issues/867
+RUN apt-get -y install locales
+RUN touch /usr/share/locale/locale.alias
+RUN sed -i -e 's/# en_US.UTF-8 UTF-8/en_US.UTF-8 UTF-8/' /etc/locale.gen && locale-gen
+ENV LANG=en_US.UTF-8
+ENV LANGUAGE=en_US:en
+ENV LC_ALL=en_US.UTF-8
+
+# set environment variables to only use one core
+RUN export OPENBLAS_NUM_THREADS=1
+RUN export MKL_NUM_THREADS=1
+RUN export BLAS_NUM_THREADS=1
+RUN export OMP_NUM_THREADS=1
+
+# install build requirements
+RUN apt install -y python3-dev python3-pip
+RUN pip3 install --upgrade setuptools
+RUN apt install -y build-essential
+
+RUN apt install -y swig
+
+# Copy the checkout autosklearn version for installation
+#ADD . /auto-sklearn/
+
+# Upgrade pip then install dependencies
+RUN pip3 install --upgrade pip
+
+# Install
+RUN pip3 install "auto-sklearn[test, examples]"
+
+COPY autosklearn_example.py autosklearn_example.py
+
+CMD ["python3", "autosklearn_example.py"]

Chapter03/automl/README.md

@@ -0,0 +1,10 @@
+# Autosklearn example
+There are known issues around installing auto-sklearn on MacOS and Windows systems so I have set this up to run in a docker container.
+
+To run this example just run the following (this assumes you have already run ```conda env create -f mlewp-chapter03.yml```):
+
+```bash
+docker build -t autosklearn .
+docker run autosklearn
+```
+

Chapter03/automl/autosklearn_example.py

@@ -2,15 +2,20 @@
 import sklearn.datasets
 import sklearn.metrics
 import autosklearn.classification
+from sklearn.datasets import load_wine
+from sklearn.model_selection import train_test_split
 
 automl = autosklearn.classification.AutoSklearnClassifier(
     time_left_for_this_task=60,
     per_run_time_limit=30
 )
 
+X, y = load_wine(return_X_y=True)
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.30, random_state=42)
+
 automl.fit(X_train, y_train, dataset_name='wine')
 
 print(automl.show_models())
 print(automl.sprint_statistics())
 predictions = automl.predict(X_test)
-sklearn.metrics.accuracy_score(y_test, predictions)
+print(sklearn.metrics.accuracy_score(y_test, predictions))

Chapter03/automl/run_autosklearn_example.sh

@@ -0,0 +1,2 @@
+docker build -t autosklearn_image .
+docker run -it autosklearn_image

Chapter03/features/feature-engineering.py

@@ -9,7 +9,6 @@
 
 # Make a 70/30 train/test split
 X_train, X_test, y_train, y_test = train_test_split(X, y,
-                                                    test_size=0.30,
                                                     test_size=0.30,
                                                     random_state=42)
 

Chapter03/hyperparameter-opt/optuna_example.py

@@ -53,5 +53,4 @@ def objective(trial, n_folds, X, y):
     study = optuna.create_study(direction='minimize')
     study.optimize(partial(objective, n_folds=n_folds, X=X_train, y=y_train), n_trials=16)
 
-    print(study.best_trial.params)
-    print(stu)
+    print(study.best_trial.params)