tf.raw_ops.QuantizeDownAndShrinkRange

robots: noindex

Convert the quantized 'input' tensor into a lower-precision 'output', using the