def main():
ext = '.nii.gz'
search_pattern = '*'
dataset = 'test'
# local
main_dir = f'/home/anatielsantos/mestrado/datasets/dissertacao/{dataset}/image/lung_extracted/teste_clahe'
model_path = '/home/anatielsantos/mestrado/models/dissertacao/unet/unet_exp1_200epc_best.h5'
# remote
# main_dir = f'/data/flavio/anatiel/datasets/dissertacao/{dataset}/image'
# model_path = '/data/flavio/anatiel/models/dissertacao/unet_500epc_last.h5'
src_dir = '{}'.format(main_dir)
dst_dir = '{}/UnetExp1PredsBest'.format(main_dir)
nproc = mp.cpu_count()
print('Num Processadores = ' + str(nproc))
model = unet()
model.load_weights(model_path)
execExecPredictByUnet(src_dir,
dst_dir,
ext,
search_pattern,
model,
reverse=False,
desc=project_name,
parallel=False)
from io import BytesIO
import skimage.io as io
from tensorflow.python.client import device_lib
#print(device_lib.list_local_devices())
import keras.backend.tensorflow_backend as K
from sklearn import metrics
from keras.preprocessing.image import *
from train import find_patches_from_slide, read_test_data_path, predict_from_model, simple_model, unet
print(
'****************************INFERENCE FILE*******************************'
)
#model = simple_model(pretrained_weights ='/data/model/u_1.h5')
model = unet(pretrained_weights='/data/model/unet555.h5')
#model = unet(pretrained_weights ='u_1.h5')
PATCH_SIZE = 256
NUM_CLASSES = 2 # not_tumor, tumor
file_handles = []
def gen_imgs_test(slide_path,
truth_path,
samples,
batch_size,
patch_size=PATCH_SIZE,
num_epoch=1,
shuffle=True):
"""This function returns a generator that