tf.nn.xw_plus_b(x, weights, biases, name=None)

tf.nn.xw_plus_b(x, weights, biases, name=None)

Computes matmul(x, weights) + biases.

Args:

  • x: a 2D tensor. Dimensions typically: batch, in_units
  • weights: a 2D tensor. Dimensions typically: in_units, out_units
  • biases: a 1D tensor. Dimensions: out_units
  • name: A name for the operation (optional). If not specified "xw_plus_b" is used.

Returns:

A 2-D Tensor computing matmul(x, weights) + biases. Dimensions typically: batch, out_units.

Defined in tensorflow/python/ops/nn_ops.py.