def automate():
number_of_zipped_files = len(get_all_files_in_directory(zipped_data_path))
number_of_raw_files = len(get_all_files_in_directory(raw_data_path))
number_of_filtered_files = len(get_all_files_in_directory(filtered_data_path))
number_of_individual_files = len(get_all_files_in_directory(individual_data_path))
if number_of_zipped_files == 0:
logger.raise_exception('There are no .zip files to work with.', logger.LogicError)
if number_of_raw_files == 0:
extract_zip_files()
if number_of_filtered_files == 0:
filter_the_stock_files(raw_data_path, filtered_data_path)
if number_of_individual_files == 0:
combined_data_frame = load_all_stock_files()
all_unique_names = combined_data_frame.symbol.values
timer = Timer()
for n in all_unique_names:
timer.start_timer()
local_data_frame = combined_data_frame[combined_data_frame.symbol == n]
if len(local_data_frame) != 0:
local_stock_path = str(str(individual_data_path) + n + '.csv')
if not os.path.isfile(local_stock_path):
local_data_frame.drop(local_data_frame.columns[0], axis=1, inplace=True)
local_data_frame['date'] = pd.to_datetime(local_data_frame['date'])
local_data_frame = local_data_frame.sort_values(by='date')
create_csv_file(local_data_frame, local_stock_path)
print('Length of local data frame: ' + str(len(local_data_frame)) + '\tLength of combined data: ' + str(len(combined_data_frame)))
combined_data_frame = combined_data_frame[combined_data_frame.symbol != n]
timer.end_timer()
timer.print_time()
def code_to_test_01():
my_list = []
my_set = set()
for x in range(10000):
my_list.append(x)
my_set.add(x)
timer = Timer()
timer.start_timer()
z = 0
for y in range(10000):
if my_list.__contains__(y):
z += y
timer.end_timer()
timer.print_time()
timer.start_timer()
z = 0
for y in range(10000):
if my_set.__contains__(y):
z += y
timer.end_timer()
timer.print_time()
split_list = splits.get_splits_for_stock(un.replace('&', '').replace('.csv', ''))
if len(split_list) != 0:
single_data_frame = perform_splits_on_stock(split_list, single_data_frame)
inner_earnings = simulation.run_simulation_for_single_stock(single_data_frame)
if inner_earnings.made_trades():
all_earnings.append(inner_earnings)
else:
bad_stocks.append(un)
# We are finished using the information for that particular stock so we will remove it from the combined data_xv frame.
all_data = all_data[all_data.symbol != un]
inner_timer.end_timer()
inner_timer.print_time()
timer.end_timer()
timer.print_time()
# Sort all the earnings by daily total profit.
all_earnings.sort(key=lambda x: x.combined_daily, reverse=False)
for earning in all_earnings:
print(earning)
logger.log('Bad stocks:')
for bd in bad_stocks:
print(bd)
timer = Timer()
#gc.disable()
for single_stock in dm.get_all_files_in_directory(dm.individual_data_path):
timer.start_timer()
single_stock_path = dm.individual_data_path + single_stock
print(single_stock_path)
df = pd.read_csv(single_stock_path, header=None, usecols=[0, 1, 2, 3, 4])
df.columns = ['date', 'open', 'high', 'low', 'close']
df['date'] = pd.to_datetime(df['date'])
split_list = splits.get_splits_for_stock(single_stock.replace('&', '').replace('.csv', ''))
if len(split_list) != 0:
df = perform_splits_on_stock(split_list, df)
profits.append(sim.run_simple_simulation_for_single_stock(single_stock, df))
timer.end_timer()
timer.print_time()
#gc.enable()
profits.sort(key=lambda x: x.total_profit, reverse=False)
for p in profits:
print(p)