Remove the input range.
tfmot.quantization.keras.remove_input_range(
model
)
Example:
model = keras.Sequential([
layers.Dense(10, activation='relu', input_shape=(100,)),
quantize_annotate_layer(layers.Dense(2, activation='sigmoid'))
])
with quantize.quantize_scope():
model = quantize_annotate_model(model)
model = quantize_apply(model)
model = remove_input_range(model)
In certain cases, a desired input range is not required if the model itself is
internally used.
Args |
model
|
A tf.keras Sequential or Functional model which has been quantized.
|
Returns |
Returns a new tf.keras model removed input range.
|