"""Utilities for ImageNet data preprocessing & prediction decoding.
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import json
import warnings
import numpy as np
from keras_applications import get_keras_submodule
backend = get_keras_submodule('backend')
keras_utils = get_keras_submodule('utils')
CLASS_INDEX = None
CLASS_INDEX_PATH = ('https://s3.amazonaws.com/deep-learning-models/'
'image-models/imagenet_class_index.json')
# Global tensor of imagenet mean for preprocessing symbolic inputs
_IMAGENET_MEAN = None
def _preprocess_numpy_input(x, data_format, mode):
"""Preprocesses a Numpy array encoding a batch of images.
# Arguments
x: Input array, 3D or 4D.
data_format: Data format of the image array.
mode: One of "caffe", "tf" or "torch".
- caffe: will convert the images from RGB to BGR,
@RM. See changes below
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os
###########
#RM
import keras
from keras_applications import get_keras_submodule
###########
backend = get_keras_submodule('backend')
engine = get_keras_submodule('engine')
layers = get_keras_submodule('layers')
models = get_keras_submodule('models')
keras_utils = get_keras_submodule('utils')
########################
#RM
#Commented ou
#from . import imagenet_utils
#from .imagenet_utils import decode_predictions
#from .imagenet_utils import _obtain_input_shape
################################################
#RM
#Added
import os
from keras import backend as K
from keras.layers import Input, Conv2D, MaxPooling2D, AveragePooling2D, concatenate, \
Concatenate, UpSampling2D, Activation
from keras.models import Model
from keras_applications import get_keras_submodule
from keras_applications.inception_resnet_v2 import InceptionResNetV2
from tqdm import tqdm
from func.config import Config
from func.group_norm import GroupNormalization
from func.se import squeeze_excite_block
layers = get_keras_submodule('layers')
backend = get_keras_submodule('backend')
CONFIG = Config()
GN_AXIS = 3
CHANNEL_AXIS = GN_AXIS
def conv_block(prev,
num_filters,
kernel=(3, 3),
strides=(1, 1),
act='relu',
prefix=None):
name = None
if prefix is not None:
name = prefix + '_conv'
