TF Constants and Variables

In TF2, tensors can be of two types: constant or variable. However, TF1 supports placeholders for its default graph execution mode. Here, we explain some properties of both constants and variables.

TF constants

We can create immutable tensors with tf.constant(), where tf is an alias for TensorFlow. To generate a constant tensor, we can write tf.constant(15). If we type tf.constant([6., 4., 3.], [5., 1., 2.]), we create a floating-point matrix with two rows and three columns.

We can apply mathematical operations, such as addition, subtraction, multiplication, division, and square, to TF constant tensors. Indexing can also be performed on constant tensors, just like on NumPy arrays.