Source code for akida_models.imagenet.model_mobilenet

#!/usr/bin/env python
# ******************************************************************************
# Copyright 2020 Brainchip Holdings Ltd.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#    http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ******************************************************************************
"""
MobileNet model definition for ImageNet classification.

MobileNet V1 is a general architecture and can be used for multiple use cases.

This specific version includes parameter options to generate a mobilenet version
compatible for Akida with:
    - overall architecture compatible with Akida (conv stride 2 replaced with
     max pool),
    - options to quantize weights and activations,
    - different initialization options.
"""

__all__ = ["mobilenet_imagenet", "mobilenet_imagenet_pretrained"]

import tensorflow as tf
from tf_keras import Model, regularizers
from tf_keras.layers import Input, Dropout, Rescaling

from .imagenet_utils import obtain_input_shape
from ..layer_blocks import conv_block, separable_conv_block, dense_block
from ..utils import fetch_file, get_params_by_version
from ..model_io import load_model, get_model_path, get_default_bitwidth


[docs] def mobilenet_imagenet(input_shape=None, alpha=1.0, dropout=1e-3, include_top=True, pooling=None, classes=1000, use_stride2=True, input_scaling=(128, -1)): """Instantiates the MobileNet architecture. Note: input preprocessing is included as part of the model (as a Rescaling layer). This model expects inputs to be float tensors of pixels with values in the [0, 255] range. Args: input_shape (tuple, optional): shape tuple. Defaults to None. alpha (float, optional): controls the width of the model. Defaults to 1.0. * If `alpha` < 1.0, proportionally decreases the number of filters in each layer. * If `alpha` > 1.0, proportionally increases the number of filters in each layer. * If `alpha` = 1, default number of filters from the paper are used at each layer. dropout (float, optional): dropout rate. Defaults to 1e-3. include_top (bool, optional): whether to include the fully-connected layer at the top of the model. Defaults to True. pooling (str, optional): optional pooling mode for feature extraction when `include_top` is `False`. Defaults to None. * `None` means that the output of the model will be the 4D tensor output of the last convolutional block. * `avg` means that global average pooling will be applied to the output of the last convolutional block, and thus the output of the model will be a 2D tensor. classes (int, optional): optional number of classes to classify images into, only to be specified if `include_top` is `True`. Defaults to 1000. use_stride2 (bool, optional): replace max pooling operations by stride 2 convolutions in layers separable 2, 4, 6 and 12. Defaults to True. input_scaling (tuple, optional): scale factor and offset to apply to inputs. Defaults to (128, -1). Note that following Akida convention, the scale factor is an integer used as a divisor. Returns: keras.Model: a Keras model for MobileNet/ImageNet. Raises: ValueError: in case of invalid input shape. """ # Model version management fused, post_relu_gap, relu_activation = get_params_by_version(relu_v2='ReLU7.5') # Define weight regularization, will apply to the first convolutional layer # and to all pointwise weights of separable convolutional layers. weight_regularizer = regularizers.l2(4e-5) # Define stride 2 or max pooling if use_stride2: sep_conv_pooling = None strides = 2 else: sep_conv_pooling = 'max' strides = 1 # Determine proper input shape and default size. if input_shape is None: default_size = 224 else: rows = input_shape[0] cols = input_shape[1] if rows == cols and rows in [128, 160, 192, 224]: default_size = rows else: default_size = 224 input_shape = obtain_input_shape(input_shape, default_size=default_size, min_size=32, include_top=include_top) rows = input_shape[0] cols = input_shape[1] img_input = Input(shape=input_shape, name="input", dtype=tf.uint8) # Use default input scaling (1, 0) if not provided scale, offset = (1, 0) if input_scaling is None else input_scaling x = Rescaling(1. / scale, offset, name="rescaling")(img_input) x = conv_block(x, filters=int(32 * alpha), name='conv_0', kernel_size=(3, 3), padding='same', use_bias=False, strides=2, add_batchnorm=True, relu_activation=relu_activation, kernel_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(64 * alpha), name='separable_1', kernel_size=(3, 3), padding='same', use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(128 * alpha), name='separable_2', kernel_size=(3, 3), padding='same', pooling=sep_conv_pooling, strides=strides, use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(128 * alpha), name='separable_3', kernel_size=(3, 3), padding='same', use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(256 * alpha), name='separable_4', kernel_size=(3, 3), padding='same', pooling=sep_conv_pooling, strides=strides, use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(256 * alpha), name='separable_5', kernel_size=(3, 3), padding='same', use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(512 * alpha), name='separable_6', kernel_size=(3, 3), padding='same', pooling=sep_conv_pooling, strides=strides, use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(512 * alpha), name='separable_7', kernel_size=(3, 3), padding='same', use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(512 * alpha), name='separable_8', kernel_size=(3, 3), padding='same', use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(512 * alpha), name='separable_9', kernel_size=(3, 3), padding='same', use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(512 * alpha), name='separable_10', kernel_size=(3, 3), padding='same', use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(512 * alpha), name='separable_11', kernel_size=(3, 3), padding='same', use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) x = separable_conv_block(x, filters=int(1024 * alpha), name='separable_12', kernel_size=(3, 3), padding='same', pooling=sep_conv_pooling, strides=strides, use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, pointwise_regularizer=weight_regularizer) # Last separable layer with global pooling layer_pooling = 'global_avg' if include_top or pooling == 'avg' else None x = separable_conv_block(x, filters=int(1024 * alpha), name='separable_13', kernel_size=(3, 3), padding='same', pooling=layer_pooling, use_bias=False, add_batchnorm=True, relu_activation=relu_activation, fused=fused, post_relu_gap=post_relu_gap, pointwise_regularizer=weight_regularizer) if include_top: x = Dropout(dropout, name='dropout')(x) x = dense_block(x, units=classes, name='classifier', use_bias=False, add_batchnorm=False, relu_activation=False, kernel_regularizer=weight_regularizer) # Create model. return Model(img_input, x, name='mobilenet_%0.2f_%s_%s' % (alpha, rows, classes))
[docs] def mobilenet_imagenet_pretrained(alpha=1.0, quantized=True, bitwidth=None): """ Helper method to retrieve a `mobilenet_imagenet` model that was trained on ImageNet dataset. Args: alpha (float): width of the model. quantized (bool, optional): a boolean indicating whether the model should be loaded quantized or not. Defaults to True. bitwidth (int, optional): the number of bits for quantized model. Defaults to None. Returns: keras.Model: a Keras Model instance. """ if bitwidth is None: bitwidth = get_default_bitwidth() if quantized and bitwidth not in [4, 8]: raise ValueError(f"bitwidth={bitwidth} is not available, use bitwidth=4 or 8 for this" " model.") if alpha == 1.0: if quantized: if bitwidth == 4: model_name_v1 = 'mobilenet_imagenet_224_iq8_wq4_aq4.h5' file_hash_v1 = '4e63ea329b3f2f773a0b9d6fef4e449639fda89d17e48bb7132b6ad86585c867' model_name_v2 = 'mobilenet_imagenet_224_alpha_1_i8_w4_a4.h5' file_hash_v2 = 'c39404d8d42e0c9c3f8ec7cd6e20e4abe2e0776e3e8e5c1ef0c76442c2f6871f' elif bitwidth == 8: model_name_v1, file_hash_v1 = None, None model_name_v2 = 'mobilenet_imagenet_224_alpha_1_i8_w8_a8.h5' file_hash_v2 = 'da5dc8433d14cc93ea6524321758b831221eea84c81d7f8709f1d20ec6ee503a' else: model_name_v1 = 'mobilenet_imagenet_224.h5' file_hash_v1 = 'fb674f627337995f4aa372e4a270445457573fc7aae61e28c05f94da14cf2980' model_name_v2 = 'mobilenet_imagenet_224_alpha_1.h5' file_hash_v2 = '23e3a9e7d5d11fd1db2ecb0caa5c909c1cadb0f3b8c72e10adce08596909b1f1' elif alpha == 0.5: if quantized: if bitwidth == 4: model_name_v1 = 'mobilenet_imagenet_224_alpha_50_iq8_wq4_aq4.h5' file_hash_v1 = 'af9a31774467de96acceeed46cee14d7864fa2cc247601beff5521e7a5e6da99' model_name_v2 = 'mobilenet_imagenet_224_alpha_0.5_i8_w4_a4.h5' file_hash_v2 = '9d0d971a76649e3d70f8f6e8aba0f74d5f773c9be560fa737bb769314c5e723c' elif bitwidth == 8: model_name_v1, file_hash_v1 = None, None model_name_v2 = 'mobilenet_imagenet_224_alpha_0.5_i8_w8_a8.h5' file_hash_v2 = 'af3e9fde35d0eeb5a69a3058ef4b0e571ef08f2ac88ae98024b134d330409131' else: model_name_v1 = 'mobilenet_imagenet_224_alpha_50.h5' file_hash_v1 = '1bffadb48f3fd194cdeecf4e02d91d0b3a87df103620c645f1c00f5e5cfbca9a' model_name_v2 = 'mobilenet_imagenet_224_alpha_0.5.h5' file_hash_v2 = '6771032d03e20afa6861610fb31621b6c85e25a283623c0fe12f77f3728fd56e' elif alpha == 0.25: if quantized: if bitwidth == 4: model_name_v1 = 'mobilenet_imagenet_224_alpha_25_iq8_wq4_aq4.h5' file_hash_v1 = '35ad51e662ed04b68978865ccb1c16bf024fa013146488b2a9c18c80d1efab29' model_name_v2 = 'mobilenet_imagenet_224_alpha_0.25_i8_w4_a4.h5' file_hash_v2 = '15fe5c0bdcca1528be312b01a76722c44987e1a4fd728a9522028e8a340e0e31' elif bitwidth == 8: model_name_v1, file_hash_v1 = None, None model_name_v2 = 'mobilenet_imagenet_224_alpha_0.25_i8_w8_a8.h5' file_hash_v2 = 'faa7155478388ca017affc561a9f2b33dcf74514c65073060803974713aad7b2' else: model_name_v1 = 'mobilenet_imagenet_224_alpha_25.h5' file_hash_v1 = '33c49bb4ec868558a2a9687be19a7e51756ddb9f00a032535069dcd5a727e0dd' model_name_v2 = 'mobilenet_imagenet_224_alpha_0.25.h5' file_hash_v2 = '8656689abb7ddabcb72fc2d230c187891d3813cdfc54d6312cfbb81970ab1d17' else: raise ValueError( f"Requested model with alpha={alpha} is not available.") model_path, model_name, file_hash = get_model_path("mobilenet", model_name_v1, file_hash_v1, model_name_v2, file_hash_v2) model_path = fetch_file(model_path, fname=model_name, file_hash=file_hash, cache_subdir='models') return load_model(model_path)