image_fullpath_to_predict = args["image_fullpath"]
image_label_real = args["image_label_real"]
input_data = []
input_labels = []
predict_flag = False
if image_fullpath_to_predict:
input_data = [image_fullpath_to_predict]
predict_flag = True
if image_label_real:
input_labels = [image_label_real]
train_set = [image_fullpath_to_predict]
setting_object = SettingsObject.Settings(
Dictionary.string_settings_german_signal_path)
option_problem = Dictionary.string_option_signals_images_problem
options = [option_problem, cv2.IMREAD_GRAYSCALE, 60, 60]
number_of_classes = 59 # Start in 0
models = models.TFModels(setting_object=setting_object,
option_problem=options,
input_data=input_data,
test=None,
input_labels=input_labels,
test_labels=None,
number_of_classes=number_of_classes,
type=None,
validation=None,
validation_labels=None,
load_model_configuration=False,
predict_flag=predict_flag)
models.convolution_model_image()
validation_set_web_traffic = tf_reader_web_traffic.validation_set # Test Set
train_set_web_traffic = tf_reader_web_traffic.train_set # Train Set
del tf_reader_web_traffic
names_of_data = [
"input_data", "validation_data", "inputs_labels", "validation_labels"
]
names_of_data_updated = [
"input_data_updated", "validation_data_updated", "inputs_labels",
"validation_labels"
]
names_dictionaries = ["input_validation_dictionary"]
# Load input, validation and labels from updated arrays where inputs are [number, float] where number is
# the page id and float is the visits' number
input_data, validation, input_labels, validation_labels = \
load_numpy_arrays_generic(path_to_load=setting_object_web_traffic.accuracies_losses_path,
names=names_of_data_updated)
models_zillow_price = models.TFModels(
input_data=input_data,
input_labels=input_labels,
validation=validation,
validation_labels=validation_labels,
number_of_classes=number_of_classes_web_traffic,
setting_object=setting_object_web_traffic,
option_problem=option_problem_web_traffic,
load_model_configuration=False)
#with tf.device('/gpu:0'):
models_zillow_price.rnn_lstm_web_traffic_time()
reader_features=reader_features) # Reader Object with all information
"""
# --------------------------------------------------------------------------
# --------------------------------------------------------------------------
# ---- DATA MINING ----
# --------------------------------------------------------------------------
# --------------------------------------------------------------------------
"""
"""
Manipulate Reader with DataMining and update it.
"""
"""
Getting train, validation (if necessary) and test set.
"""
test_set = tf_reader.test_set # Test Set
train_set = tf_reader.train_set # Train Set
del reader_features
del tf_reader
option_problem = Dictionary.string_option_signals_images_problem
models = models.TFModels(input=train_set[0],
test=test_set[0],
input_labels=train_set[1],
test_labels=test_set[1],
number_of_classes=number_of_classes,
setting_object=setting_object,
option_problem=option_problem,
load_model_configuration=True)
models.convolution_model_image()
# --------------------------------------------------------------------------
# ---- DATA MINING ----
# --------------------------------------------------------------------------
# --------------------------------------------------------------------------
"""
"""
Manipulate Reader with DataMining and update it.
"""
"""
Getting train, validation (if necessary) and test set.
"""
train_set = tf_reader.train_set # Train Set
test_set = tf_reader.test_set # Test Set
del reader_features
del tf_reader
models = models.TFModels(setting_object=setting_object,
option_problem=options,
input_data=train_set[0],
test=test_set[0],
input_labels=train_set[1],
test_labels=test_set[1],
number_of_classes=number_of_classes,
type=None,
validation=None,
validation_labels=None,
load_model_configuration=False)
#with tf.device('/cpu:0'): # CPU
with tf.device('/gpu:0'): # GPU
models.convolution_model_image()
reader_features=reader_features,
settings=setting_object) # Reader Object with all information
x_train = tf_reader.x_train
y_train = tf_reader.y_train
x_test = tf_reader.x_test
y_test = tf_reader.y_test
pt("x_train", x_train.shape)
pt("y_train", y_train.shape)
pt("x_test", x_test.shape)
pt("y_test", y_test.shape)
with tf.device('/gpu:0'): # GPU
models = models.TFModels(setting_object=setting_object,
option_problem=options,
input_data=x_train,
test=x_test,
input_labels=y_train,
test_labels=y_test,
number_of_classes=number_of_classes,
type=None,
validation=None,
validation_labels=None)
#with tf.device('/cpu:0'): # CPU
models.convolution_model_image()
"""
if __name__ == '__main__':
import multiprocessing
multiprocessing.freeze_support()
"""