def main():
# read data from csv to data frame
data_arr = np.genfromtxt(sys.argv[1], delimiter=',', dtype="|U5")
# applies one hot enconding over samples set
samples_size = len(data_arr) # number of samples in training set
data_arr = one_hot_encoding(data_arr, samples_size)
# read labels training set data
test_x = np.genfromtxt(sys.argv[3], delimiter=',', dtype="|U5")
test_size = len(test_x) # number of samples in test set
test_x = one_hot_encoding(test_x, test_size)
test_x = np.array(test_x, dtype=float)
test_x = z_score(test_x)
# normalize the data-set
data_set = np.array(data_arr, dtype=float)
data_set = z_score(data_set)
# load labels file
label_set = np.genfromtxt(sys.argv[2], delimiter=',', dtype=float)
# split the data set to 80% training set and 20% to the test set
# trained models with perceptron, svm and passive aggressive algorithm
regulation_constant = 0.25 # regulation constant value
m_perc = perceptron.getBestModel(data_set, label_set, samples_size)
m_svm = svm.getBestModel(data_set, label_set, samples_size,
regulation_constant)
m_pa = pa.getBestModel(data_set, label_set, samples_size)
# prints out the prediction of given models for each sample in test set
tester.print_results(m_perc, m_svm, m_pa, test_x, test_size)
[test_create_order, True],
[test_get_all_orders, True],
[test_get_requested_orders, True],
[test_get_requested_and_cooking_orders, True],
]
valid_order_id = None
def get_valid_order_id():
global valid_order_id
if valid_order_id:
return valid_order_id
valid_order_req = session.post(api_url + "create_order",
json={
"table_num": 1,
"items": [1]
})
valid_order_id = valid_order_req.json().get("data")["order_id"]
return valid_order_id
if __name__ == "__main__":
if len(sys.argv) > 1 and sys.argv[1] == "v":
verbose = True
total = 0
passed = 0
failed = []
total, passed, failed = tester.run_tests(tests, total, passed, failed)
tester.print_results(total, passed, failed)