potential TensorFlow benchmark improvements #91

Hi -- thanks for the benchmarks!

I noticed that you do a sparse-to-dense conversion and use softmax_cross_entropy_with_logits. Have you tried eliding the sparse-to-dense conversion and using sparse_softmax_cross_entropy_with_logits? In my experience the sparse version is faster.

Also, reduce_mean does not have a GPU kernel. reduce_sum with a division would prevent a GPU -> CPU -> GPU step when calculating the loss.