Quantizes then dequantizes a tensor.
tf.raw_ops.QuantizeAndDequantizeV3(
input, input_min, input_max, num_bits, signed_input=True, range_given=True,
narrow_range=False, axis=-1, name=None
)
This is almost identical to QuantizeAndDequantizeV2, except that num_bits is a tensor, so its value can change during training.
Args | |
---|---|
input
|
A Tensor . Must be one of the following types: bfloat16 , half , float32 , float64 .
|
input_min
|
A Tensor . Must have the same type as input .
|
input_max
|
A Tensor . Must have the same type as input .
|
num_bits
|
A Tensor of type int32 .
|
signed_input
|
An optional bool . Defaults to True .
|
range_given
|
An optional bool . Defaults to True .
|
narrow_range
|
An optional bool . Defaults to False .
|
axis
|
An optional int . Defaults to -1 .
|
name
|
A name for the operation (optional). |
Returns | |
---|---|
A Tensor . Has the same type as input .
|