Configure setup and Travis; also remove Cleverhans from config

Author: Irina Nicolae

.travis.yml

@@ -0,0 +1,61 @@
+sudo: required
+dist: trusty
+language: python
+python:
+  - "2.7"
+  - "3.5"
+matrix:
+  include:
+      - python: 2.7
+        env:
+          - KERAS_BACKEND=tensorflow
+          - TENSORFLOW_V=1.4.1
+      - python: 2.7
+        env:
+          - KERAS_BACKEND=tensorflow
+          - TENSORFLOW_V=1.1.0
+      - python: 3.5
+        env:
+          - KERAS_BACKEND=tensorflow
+          - TENSORFLOW_V=1.1.0
+      - python: 3.5
+        env:
+          - KERAS_BACKEND=tensorflow
+          - TENSORFLOW_V=1.4.1
+
+before_install:
+  - if [[ "$TRAVIS_PYTHON_VERSION" == "2.7" ]]; then
+      wget https://repo.continuum.io/miniconda/Miniconda2-latest-Linux-x86_64.sh -O miniconda.sh;
+    else
+      wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh;
+    fi
+  - bash miniconda.sh -b -p $HOME/miniconda
+  - export PATH="$HOME/miniconda/bin:$PATH"
+  - hash -r
+  - conda config --set always_yes yes --set changeps1 no
+  - conda update -q conda
+  - conda info -a
+
+install:
+- conda create -q -n test-environment python=$TRAVIS_PYTHON_VERSION numpy scipy matplotlib pandas h5py
+- source activate test-environment
+  - conda create -q -n test-environment python=$TRAVIS_PYTHON_VERSION
+  - source activate test-environment
+  # Install TensorFlow
+  - if [[ "$TRAVIS_PYTHON_VERSION" == "2.7" && "$TENSORFLOW_V" == "1.4.1" ]]; then
+        pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.4.1-cp27-none-linux_x86_64.whl;
+      elif [[ "$TRAVIS_PYTHON_VERSION" == "2.7" && "$TENSORFLOW_V" == "1.1.0" ]]; then
+        pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.1.0-cp27-none-linux_x86_64.whl;
+      elif [[ "$TRAVIS_PYTHON_VERSION" == "3.5" && "$TENSORFLOW_V" == "1.4.1" ]]; then
+        pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.4.1-cp35-cp35m-linux_x86_64.whl;
+      elif [[ "$TRAVIS_PYTHON_VERSION" == "3.5" && "$TENSORFLOW_V" == "1.1.0" ]]; then
+        pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.1.0-cp35-cp35m-linux_x86_64.whl;
+  fi
+  - pip install keras
+  # Run setup
+  - conda install libgcc
+  - export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/home/travis/miniconda/envs/test-environment/lib
+
+script:
+  - mkdir ./data
+  - python -m unittest discover src/ -p '*_unittest.py'

README.md

@@ -12,51 +12,47 @@ Nemesis contains implementations of the following attacks:
 * Universal Perturbation ([Moosavi-Dezfooli et al., 2016](https://arxiv.org/abs/1610.08401))
 * Virtual Adversarial Method ([Moosavi-Dezfooli et al., 2015](https://arxiv.org/abs/1507.00677))
 * C&W Attack ([Carlini and Wagner, 2016](https://arxiv.org/abs/1608.04644))
+* NewtonFool ([Jang et al., 2017](http://doi.acm.org/10.1145/3134600.3134635))
 
 The following defense methods are also supported:
 * Feature squeezing ([Xu et al., 2017](http://arxiv.org/abs/1704.01155))
+* Spatial smoothing ([Xu et al., 2017](http://arxiv.org/abs/1704.01155))
 * Label smoothing (Warde-Farley and Goodfellow, 2016)
 * Adversarial training ([Szegedy et al., 2013](http://arxiv.org/abs/1312.6199))
 * Virtual adversarial training ([Miyato et al., 2017](https://arxiv.org/abs/1704.03976))
 
 ## Setup
 
-### Requirements
-
 Nemesis is designed to run with Python 3 (and most likely Python 2 with small changes). You can either download the source code of Nemesis or clone the repository in your directory of choice:
 ```bash
 git clone https://github.ibm.com/Maria-Irina-Nicolae/nemesis
 ```
 
 To install the project dependencies, use the requirements file:
 ```bash
-pip install -r requirements.txt
+pip install .
 ```
 
-You will additionally need to download [Cleverhans](https://github.com/tensorflow/cleverhans).
-
-### Installation
+The library comes with a basic set of unit tests. To check your install, you can run all the unit tests by calling in the Nemesis folder:
+```bash
+bash run_tests.sh
+```
 
-Nemesis is linked against Cleverhans through the configuration file `config/config.ini`. When installing Nemesis on your local machine, you need to set the appropriate paths and the `LOCAL` configuration profile as follows:
+The configuration file `config/config.ini` allows to set custom paths for data. By default, data is downloaded in the `nemesis/data` folder as follows:
 
 ```text
 [DEFAULT]
 profile=LOCAL
 
 [LOCAL]
-data_path=/local/path/here
-mnist_path=/local/path/here
-cifar10_path=/local/path/here
-stl10_path=/local/path/here
-cleverhans_path=/local/path/here
+data_path=./data
+mnist_path=./data/mnist
+cifar10_path=./data/cifar-10
+stl10_path=./data/stl-10
 ```
 
 If the datasets are not present at the indicated path, loading them will also download the data.
 
-The library comes with a basic set of unit tests. To check that the installation has succeeded, you can run all the unit tests by calling in the Nemesis folder:
-```bash
-bash run_tests.sh
-```
 
 ## Running Nemesis
 

config/config.ini

@@ -1,18 +1,16 @@
 [DEFAULT]
-profile=CLUSTER
+profile=LOCAL
 
 [LOCAL]
-data_path=/local/path/here
-mnist_path=/local/path/here
-cifar10_path=/local/path/here
-stl10_path=/local/path/here
-imagenet_path=/local/path/here
-cleverhans_path=/local/path/here
+data_path=./data
+mnist_path=./data/mnist
+cifar10_path=./data/cifar-10
+stl10_path=./data/stl-10
+imagenet_path=./data/imagenet
 
 [CLUSTER]
 data_path=/dccstor/dlw/data/adversarial_learning
 mnist_path=/dccstor/dlw/data/adversarial_learning/mnist
 cifar10_path=/dccstor/dlw/data/adversarial_learning/cifar-10
 stl10_path=/dccstor/dlw/data/adversarial_learning/stl-10
 imagenet_path=/dccstor/dlw/data/kg_completion/images/jpegs_ILSVRC/n02119789
-cleverhans_path=/u/nicolae/cleverhans

requirements.txt

@@ -1,4 +1,5 @@
-Keras==2.0.4
-matplotlib==2.0.1
-scipy==0.19.0
-tensorflow==1.1.0
+h5py
+Keras
+matplotlib
+scipy
+tensorflow

setup.py

@@ -0,0 +1,36 @@
+from setuptools import setup
+from setuptools import find_packages
+
+
+setup(name='Adversarial Robustness Toolbox',
+      version='0.1',
+      description='IBM Adversarial machine learning toolbox',
+      author='Irina Nicolae',
+      author_email='[email protected]',
+      url='https://github.com/ririnicolae/adversarial-robustness-toolbox',
+      license='MIT',
+      install_requires=['h5py',
+                        'Keras',
+                        'scipy',
+                        'matplotlib',
+                        'tensorflow',
+                        'setuptools'],
+      # extras_require={
+          # 'tests': ['pytest',
+          #           'pytest-pep8',
+          #           'pytest-xdist',
+          #           'pytest-cov'],
+      # },
+      classifiers=[
+            'Development Status :: 3 - Alpha',
+            'Intended Audience :: Developers',
+            'Intended Audience :: Education',
+            'Intended Audience :: Science/Research',
+            'License :: OSI Approved :: MIT License',
+            'Programming Language :: Python :: 2',
+            'Programming Language :: Python :: 3',
+            'Topic :: Software Development :: Libraries',
+            'Topic :: Software Development :: Libraries :: Python Modules',
+            'Topic :: Scientific/Engineering :: Artificial Intelligence',
+      ],
+      packages=find_packages())