Merge pull request #9 from tgsmith61591/release-0.4

Release 0.4

.coveragerc

@@ -1,5 +1,6 @@
 [run]
 source = smrt
+include = */smrt/*
 omit =
     */python?.?/*
     */lib-python/?.?/*.py

.gitignore

@@ -35,6 +35,9 @@ var/
 .installed.cfg
 *.egg
 
+# Serialized python
+*.pkl
+
 # PyInstaller
 #  Usually these files are written by a python script from a template
 #  before PyInstaller builds the exe, so as to inject date/other infos into it.

README.md

@@ -1,13 +1,14 @@
 [![Build status](https://travis-ci.org/tgsmith61591/smrt.svg?branch=master)](https://travis-ci.org/tgsmith61591/smrt)
-![Supported versions](https://img.shields.io/badge/python-2.7-blue.svg) 
+[![codecov](https://codecov.io/gh/tgsmith61591/smrt/branch/master/graph/badge.svg)](https://codecov.io/gh/tgsmith61591/smrt)
+![Supported versions](https://img.shields.io/badge/python-2.7-blue.svg)
 ![Supported versions](https://img.shields.io/badge/python-3.5-blue.svg)
 
 
-# Synthetic Minority Reconstruction Technique (SMRT)
+# Synthetic Minority Reconstruction Technique ([SMRT](https://www.youtube.com/watch?v=tcGQpjCztgA))
 *Handle your class imbalance more intelligently by using SMOTE's younger, more sophisticated cousin*
 
 
-### Installation
+## Installation
 
 Installation is easy. After cloning the project onto your machine and installing the required dependencies,
 simply use the `setup.py` file:
@@ -18,7 +19,7 @@ $ cd smrt
 $ python setup.py install
 ```
 
-### About
+## About
 
 SMRT (Sythetic Minority Reconstruction Technique) is the new SMOTE (Synthetic Minority Oversampling TEchnique).
 Using variational auto-encoders, SMRT learns the latent factors that best reconstruct the observations in each
@@ -32,8 +33,25 @@ too close to the decision boundary for reliable classification, and could lead t
 of an estimator. SMRT avoids this risk implicitly, as the [``VariationalAutoencoder``](smrt/autoencode/autoencoder.py)
 learns a distribution that is generalizable to the lowest-error (i.e., most archetypal) observations.
 
+__See [the paper](doc/smrt.tex) for more in-depth reference.__
+
+## Example
+
+The [SMRT example](examples/) is an ipython notebook with reproducible code and data that compares an imbalanced
+variant of the MNIST dataset after being balanced with both SMOTE and SMRT. The following are several of the resulting
+images produced from both SMOTE and SMRT, respectively. Even visually, it's evident that SMRT better synthesizes data
+that resembles the input data.
+
+### Original:
+
+The MNIST dataset was amended to contain only zeros and ones in an unbalanced (~1:100, respectively) ratio. The top row
+are the original MNIST images, the second row is the SMRT-generated images, and the bottom row is the SMOTE-generated
+images:
+<br/>
+<img src="examples/img/mnist_smrt_smote.png" width="600" alt="Original"/>
+
 ### Notes
 
 - See [examples](examples/) for usage
+- See [the paper](doc/smrt.tex) for more in-depth documentation
 - Information on [the authors](AUTHORS.md)
-- S-M-R-T [original reference](https://www.youtube.com/watch?v=tcGQpjCztgA) (enable audio)

build_tools/travis/install.sh

@@ -43,13 +43,15 @@ if [[ "$DISTRIB" == "conda" ]]; then
 
     pip install scikit-learn==$SCIKIT_LEARN_VERSION
     pip install coveralls
+    pip install codecov
 
     # Install TensorFlow
     # we have to make sure we install the CPU version otherwise we get into GCC/G++ issues...
     if [[ "$PYTHON_VERSION" == "2.7" ]]; then
       pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp27-none-linux_x86_64.whl;
     elif [[ "$PYTHON_VERSION" == "3.5" ]]; then
-      pip install http://ci.tensorflow.org/view/Nightly/job/nightly-python35-linux-cpu/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.0rc2-cp35-cp35m-linux_x86_64.whl;
+      # pip install http://ci.tensorflow.org/view/Nightly/job/nightly-python35-linux-cpu/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.0rc2-cp35-cp35m-linux_x86_64.whl;
+      pip install https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0-cp35-cp35m-linux_x86_64.whl;
     fi
 
     # Install nose-timer via pip

build_tools/travis/test_script.sh

@@ -14,7 +14,7 @@ python -c "import scipy; print('scipy %s' % scipy.__version__)"
 python -c "import multiprocessing as mp; print('%d CPUs' % mp.cpu_count())"
 
 run_tests() {
-    # Get into a temp directory to run test from the installed scikit learn and
+    # Get into a temp directory to run test from the installed smrt and
     # check if we do not leave artifacts
     mkdir -p $TEST_DIR
     # We need the setup.cfg for the nose settings
@@ -23,6 +23,7 @@ run_tests() {
 
     if [[ "$COVERAGE" == "true" ]]; then
         nosetests -s --with-coverage --with-timer --timer-top-n 20 smrt
+        codecov --token=6de72173-c50f-4ebb-b7f1-22d592f56b01
     else
         nosetests -s --with-timer --timer-top-n 20 smrt
     fi