def setUpClass(cls):
cls.train_dataset, cls.train_labels, cls.valid_dataset, cls.valid_labels, cls.test_dataset, cls.test_labels = get_data_4d()
dataholder = DataHolder(cls.train_dataset,
cls.train_labels,
cls.valid_dataset,
cls.valid_labels,
cls.test_dataset,
cls.test_labels)
config = Config()
cls.model = CNNModel(config, dataholder)
import numpy as np
import inspect
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
currentdir = os.path.dirname(os.path.abspath(inspect.getfile(inspect.currentframe())))
parentdir = os.path.dirname(currentdir)
sys.path.insert(0, parentdir)
from util import run_test, get_data_4d, get_time
from CNN import CNNModel, train_model, check_valid
from DataHolder import DataHolder
from Config import Config
train_dataset, train_labels, valid_dataset, valid_labels, test_dataset, test_labels = get_data_4d()
my_dataholder = DataHolder(train_dataset,
train_labels,
valid_dataset,
valid_labels,
test_dataset,
test_labels)
number_of_exp = 10
DECAY = np.random.random_sample([number_of_exp])
DECAY = np.append(DECAY, 0.96)
number_of_exp += 1
DECAY.sort()
results = []
duration = []
def main():
"""
Basic script that shows the training of the model,
the accuracy of the test and valid datasets, and
the prediction of one specific image.
"""
train_dataset, train_labels, valid_dataset, valid_labels, test_dataset, test_labels = get_data_4d(
)
# my_config = Config()
my_config = Config(learning_rate=0.52660658241,
dropout=0.75,
batch_size=230,
steps_for_decay=800,
num_filters_1=12,
hidden_nodes_3=300,
hidden_nodes_1=900,
decay_rate=0.349144998004,
num_filters_2=24,
patch_size=11,
image_size=28,
hidden_nodes_2=600)
my_dataholder = DataHolder(train_dataset, train_labels, valid_dataset,
valid_labels, test_dataset, test_labels)
my_model = CNNModel(my_config, my_dataholder)
train_model(my_model, my_dataholder, 4 * 10001, 1000)
print("check_valid = ", check_valid(my_model))
print("check_test = ", check_test(my_model))
one_example = test_dataset[0]
one_example = one_example.reshape(1, one_example.shape[0],
one_example.shape[1],
one_example.shape[2])
prediction = chr(one_prediction(my_model, one_example) + ord('A'))
real = chr(np.argmax(test_labels[0]) + ord('A'))
print("Prediction = {}".format(prediction))
print("Real label = {}".format(real))
