def run(travel_search_date):
report = {
"date": datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S'),
"travel_search_date": travel_search_date,
}
if stop_times_file is not None:
stops = csv_loader.load_csv(stop_times_file)
print("loaded {number_of_searches} stops from file".format(
number_of_searches=len(stops)))
report["stopTimes"] = stop_times_executor.run_stop_times_searches(
stops)
travel_searches = csv_loader.load_csv(travel_search_file)
print("loaded {number_of_searches} searches from file".format(
number_of_searches=len(travel_searches)))
print("Running searches against endpoint " + graphql_endpoint)
report["travelSearch"] = travel_search_executor.run_travel_searches(
travel_searches, travel_search_date)
json_report = json.dumps(report)
filename = report_dao.save_json_report(json_report)
if upload_gcp:
gcp_uploader.upload_blob(os.environ[BUCKET_NAME_ENV], filename,
os.environ[DESTINATION_BLOB_NAME_ENV])
# Consider using constructor for gcp uploader
gcp_uploader.remove_old_files(os.environ[BUCKET_NAME_ENV],
os.environ[DESTINATION_BLOB_NAME_ENV])
def main():
fname = sys.argv[1]
sys.argv.pop(1)
csv = np.loadtxt(fname, dtype='str', delimiter=',')
data = csv_loader.load_csv(csv, fname)
closest_row = data.find_closest_row(1481747621, 281100000)
argCount = len(sys.argv)
if argCount != 4:
print "Wrong number of arguments"
exit(1)
# SET_NAME_ID = "M19"
# SET_NAME = "Core Set 2019"
# CSV_FILENAME = "c:\\card\\csv\M19_Wanted.csv"
SET_NAME_ID = sys.argv[1] # XLN, M19, etc
SET_NAME = sys.argv[2] # Ixalan, Core Set 19
CSV_FILENAME = sys.argv[3] # "c:\\card\\csv\M19_Wanted.csv"
card_name_array = csv_loader.load_csv(CSV_FILENAME)
all_cards = []
# add_price_to_csv(CSV_FILENAME, all_cards)
for card in card_name_array:
card_description = get_card_description(card.strip(), SET_NAME_ID,
SET_NAME, False)
if card_description is None:
print "ERROR : can't find card %s" % card.strip()
else:
all_cards.append(card_description)
browser.close()
sorted_cards = sorted(all_cards, key=lambda k: float(k["tix"]))
def load_csv(filename):
return csv_loader.load_csv(filename)
# Build a search urls for cardhoarder.com
def build_mtgo_url(set_name_id, card_name):
clean_card_name = card_name.replace(' ', '_')
clean_card_name = clean_card_name.replace('\'', '')
clean_card_name = clean_card_name.replace(',', '')
url = "https://www.mtgotraders.com/store/" + set_name_id + "_" + clean_card_name + ".html"
return url
browser = webdriver.Chrome("c:\\workspace\\python\\chromedriver.exe")
SET_NAME_ID = "M19"
SET_NAME = "Core Set 2019"
card_name_array = csv_loader.load_csv("c:\\card\\csv\M19_Wanted.csv")
# card_name_array = ["Stitcher's Supplier"]
all_cards = []
for card in card_name_array:
url = build_mtgo_url(SET_NAME_ID, card)
html = download_webpage(url)
price = get_price_from_webpage(html)
card_description = {}
card_description['name'] = card
card_description['price'] = float(price)
all_cards.append(card_description)
sorted_cards = sorted(all_cards, key=lambda k: k["price"])
# gradients at last layer
dW2 = (1 / m_batch) * np.matmul(dZ2, cache["A1"].T)
db2 = (1. / m_batch) * np.sum(dZ2, axis=1, keepdims=True)
# back propagate trough first layer
dA1 = np.matmul(params["W2"].T, dZ2)
dZ1 = dA1 * sigmoid(cache["Z1"]) * (1 - sigmoid(cache["Z1"]))
# gradients at first layer
dW1 = (1 / m_batch) * np.matmul(dZ1, X.T)
db1 = (1 / m_batch) * np.sum(dZ1, axis=1, keepdims=True)
return {"dW1": dW1, "db1": db1, "dW2": dW2, "db2": db2}
train_images, train_labels, test_images = csv_loader.load_csv()
X_train, X_test = train_images.T / 255, test_images.T / 255 # (784, 60000), (784, 10000)
# one-hot encoding
digits = 10
Y_train = np.eye(digits)[train_labels.T.astype('int32')]
Y_train = Y_train.T.reshape(digits, train_labels.shape[0]) # (10, 60000)
# Shuffle the training set
# np.random.seed(138)
# shuffle_index = np.random.permutation(X_train.shape[1])
# X_train, Y_train = X_train[:, shuffle_index], Y_train[:, shuffle_index]
# hyperparameters
n_1 = X_train.shape[0] # 784
n_2 = 64