This is a simplified interface for TensorFlow, to get people started on predictive analytics and data mining.
First, make sure you have TensorFlow and Scikit Learn installed, then just run:
pip install git+git://github.com/google/skflow.git
Example usage:
iris = datasets.load_iris()
classifier = skflow.TensorFlowClassifier(n_classes=3)
classifier.fit(iris.data, iris.target)
score = accuracy_score(classifier.predict(iris.data), iris.target)