What is the difference between np.mean and tf.reduce_mean?

Question

In the MNIST beginner tutorial, there is the statement

accuracy = tf.reduce_mean(tf.cast(correct_prediction, \"float\"))

tf.cast

Answer 1

1 usually refers to rows, and 2 usually refers to columns. Reducing "over" index 1 means to reduce rowwise.

[1., 2.] is just [ , ].

This index numbering convention is typical in stats software, especially R.