def update_with_query_conf(user_args):
"""
Update the default YAML configuration with the user's input args from the API query
"""
# Update the default conf with the user input
CONF = config.CONF
for group, val in sorted(CONF.items()):
for g_key, g_val in sorted(val.items()):
if g_key in user_args:
g_val['value'] = json.loads(user_args[g_key])
# Check and save the configuration
config.check_conf(conf=CONF)
config.conf_dict = config.get_conf_dict(conf=CONF)
def update_with_saved_conf(saved_conf):
"""
Update the default YAML configuration with the configuration saved from training
"""
# Update the default conf with the user input
CONF = config.CONF
for group, val in sorted(CONF.items()):
if group in saved_conf.keys():
for g_key, g_val in sorted(val.items()):
if g_key in saved_conf[group].keys():
g_val['value'] = saved_conf[group][g_key]
# Check and save the configuration
config.check_conf(conf=CONF)
config.conf_dict = config.get_conf_dict(conf=CONF)
def save_default_imagenet_model():
"""
Create a model in models_dir with default ImageNet training
"""
CONF = config.get_conf_dict()
TIMESTAMP = 'default_imagenet'
# Clear default conf and create custom conf
for k, v in CONF.items():
if k in ['general', 'augmentation']:
continue
for i, j in v.items():
CONF[k][i] = None
CONF['augmentation']['train_mode'] = None
CONF['model']['modelname'] = 'Xception'
CONF['model']['image_size'] = 224
CONF['model']['preprocess_mode'] = model_modes[CONF['model']['modelname']]
CONF['model']['num_classes'] = 1000
CONF['dataset']['mean_RGB'] = [123.675, 116.28, 103.53]
CONF['dataset']['std_RGB'] = [58.395, 57.12, 57.375]
paths.timestamp = TIMESTAMP
paths.CONF = CONF
# Create classes.txt for ImageNet
fpath = keras.utils.get_file(
'imagenet_class_index.json',
'https://s3.amazonaws.com/deep-learning-models/image-models/imagenet_class_index.json',
cache_subdir='models',
file_hash='c2c37ea517e94d9795004a39431a14cb')
with open(fpath) as f:
classes = json.load(f)
classes = np.array(list(classes.values()))[:, 1]
# Create the model
architecture = getattr(applications, CONF['model']['modelname'])
img_width, img_height = CONF['model']['image_size'], CONF['model']['image_size']
model = architecture(weights='imagenet', include_top=True, input_shape=(img_width, img_height, 3))
model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])
# Save everything
utils.create_dir_tree()
np.savetxt(os.path.join(paths.get_ts_splits_dir(), 'classes.txt'), classes, fmt='%s', delimiter='/n')
save_conf(CONF)
model.save(fpath=os.path.join(paths.get_checkpoints_dir(), 'final_model.h5'),
include_optimizer=False)
stats = {'epoch': history.epoch,
'training time (s)': round(time.time()-t0, 2),
'timestamp': TIMESTAMP}
stats.update(history.history)
stats = json_friendly(stats)
stats_dir = paths.get_stats_dir()
with open(os.path.join(stats_dir, 'stats.json'), 'w') as outfile:
json.dump(stats, outfile, sort_keys=True, indent=4)
print('Saving the configuration ...')
model_utils.save_conf(CONF)
print('Saving the model to h5...')
fpath = os.path.join(paths.get_checkpoints_dir(), 'final_model.h5')
model.save(fpath,
include_optimizer=False)
# print('Saving the model to protobuf...')
# fpath = os.path.join(paths.get_checkpoints_dir(), 'final_model.proto')
# model_utils.save_to_pb(model, fpath)
print('Finished')
if __name__ == '__main__':
CONF = config.get_conf_dict()
timestamp = datetime.now().strftime('%Y-%m-%d_%H%M%S')
train_fn(TIMESTAMP=timestamp, CONF=CONF)
#Importamos todo lo necesario como en el jupyter 1.0 de Ignacio
import os
import matplotlib.pylab as plt
import numpy as np
from tqdm import tqdm
import imgclas
from imgclas import paths, config
from imgclas.data_utils import load_image, load_data_splits, augment, load_class_names
#Comenzamos a preparar todos los datos
CONF = config.get_conf_dict(
) #El diccionario con toda la configuracion del yaml
splits_dir = paths.get_splits_dir() #base+data+dataset_files
# Load the training data
X_train, y_train = load_data_splits(splits_dir=splits_dir,
im_dir=CONF['general']['images_directory'],
split_name='train')
# Load the validation data
if (CONF['training']['use_validation']) and ('val.txt'
in os.listdir(splits_dir)):
X_val, y_val = load_data_splits(splits_dir=splits_dir,
im_dir=CONF['general']['images_directory'],
split_name='val')
#load_data_splits comprueba que exista el fichero que se le pasa (ya sean train,val etc). luego con numpy.genfromtxt
#obtiene un array donde la primera columna son los path, en la segunda las etiquetas
#por ultimo retorna un array de numpy con los path absolutos a las fotografias de train o el que le hayas pasado
#y otro con las etiquetas en formato int32 para saber de qué clase son
