TensorFlow - Tensors #2

In this tutorial we discuss TensorFlow basics and we take a deep dive into tesnors and tensor operations. In the math behind it, we discuss standard deviation, variance and dot product of matrices.

Github:

Twitter:

T-Shirt:

[GiuseppeRomagnuolo]

It is a really clear introduction to Tensorflow interspersed with useful hints and heads-ups, thank you.

[maxinteltech]

I like this method of teaching by interpreting codes

[froggenfury]

i cant thank you enough, you're a god!

[sungchulyonseiackr]

I like your talk a lot. Great job and many thanks!!!

However, you misrepresent the matrix shape and matrix multiplication.
You need following changes in the presentation just above the cell 40:

\mathbf{A}_{(2, 3)} ---> \mathbf{A}_{(3, 2)}

\mathbf{B}_{(3, 2)} ---> \mathbf{B}_{(2, 3)}

\mathbf{C}_{(2, 2)} = \mathbf{A}_{(2, 3)} . \mathbf{B}_{(3, 2)} ---> \mathbf{C}_{(3, 3)} = \mathbf{A}_{(3, 2)} . \mathbf{B}_{(2, 3)}

\mathbf{C}_{(2, 2)} ---> \mathbf{C}_{(3, 3)} = array([[ 2, 1, 2],
[ 8, 7, 12],
[ 6, 9, 14]], dtype=int32)


So, in cell 41 you should change:

c = tf.matmul(tf.matrix_transpose(a), tf.matrix_transpose(b))
--->
c = tf.matmul(a, b)

By the way, how do you type $\sigma$ in cell 35?