Deep Deterministic Policy Gradients in TensorFlow