def assemble_last_signals_function(config: dict):
meta_file = os.path.join("output", "metadata.json")
if not os.path.exists(meta_file):
print(
f"{WARNING}Warning: '{meta_file}' file does not exist. Run main program.{NORMAL}")
return
with open(meta_file) as m_file:
m_data = json.load(m_file)
m_file.close()
for fund in m_data:
if fund != '_METRICS_':
print("")
assemble_last_signals(
m_data[fund], standalone=True, print_out=True, name=fund)
print("")
def run_dev(script: list):
"""Run Development Script
Script that is for implementing new content
Arguments:
script {list} -- dataset, funds, periods, config
Returns:
dict -- analysis object of fund data
"""
dataset = script[0]
funds = script[1]
periods = script[2]
config = script[3]
# Start of automated process
analysis = {}
clock = start_clock()
for fund_name in funds:
if fund_name in SKIP_INDEXES:
continue
fund_print = INDEXES.get(fund_name, fund_name)
print("")
print(f"~~{fund_print}~~")
create_sub_temp_dir(fund_name, sub_periods=config['period'])
analysis[fund_name] = {}
analysis[fund_name]['metadata'] = get_api_metadata(
fund_name,
max_close=max(dataset[periods[0]][fund_name]['Close']),
data=dataset[periods[0]][fund_name])
###################### START OF PERIOD LOOPING #############################
for i, period in enumerate(periods):
fund_data = {}
fund = dataset[period][fund_name]
start = date_extractor(fund.index[0], _format='str')
end = date_extractor(fund.index[-1], _format='str')
fund_data['dates_covered'] = {'start': str(start), 'end': str(end)}
fund_data['name'] = fund_name
fund_print2 = fund_print + f" ({period}) "
p = ProgressBar(config['process_steps'],
name=fund_print2,
offset=clock)
p.start()
fund_data['statistics'] = get_high_level_stats(fund)
fund_data['clustered_osc'] = cluster_oscs(fund,
function='all',
filter_thresh=3,
name=fund_name,
plot_output=False,
progress_bar=p,
view=period)
fund_data['full_stochastic'] = full_stochastic(fund,
name=fund_name,
plot_output=False,
out_suppress=False,
progress_bar=p,
view=period)
fund_data['rsi'] = RSI(fund,
name=fund_name,
plot_output=False,
out_suppress=False,
progress_bar=p,
view=period)
fund_data['ultimate'] = ultimate_oscillator(fund,
name=fund_name,
plot_output=False,
out_suppress=False,
progress_bar=p,
view=period)
fund_data['awesome'] = awesome_oscillator(fund,
name=fund_name,
plot_output=False,
progress_bar=p,
view=period)
fund_data['momentum_oscillator'] = momentum_oscillator(
fund,
name=fund_name,
plot_output=False,
progress_bar=p,
view=period)
fund_data['on_balance_volume'] = on_balance_volume(
fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['simple_moving_average'] = triple_moving_average(
fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['exp_moving_average'] = triple_exp_mov_average(
fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['sma_swing_trade'] = moving_average_swing_trade(
fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['ema_swing_trade'] = moving_average_swing_trade(
fund,
function='ema',
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['hull_moving_average'] = hull_moving_average(
fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['macd'] = mov_avg_convergence_divergence(
fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['bear_bull_power'] = bear_bull_power(fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['total_power'] = total_power(fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['bollinger_bands'] = bollinger_bands(fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['commodity_channels'] = commodity_channel_index(
fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['rate_of_change'] = rate_of_change_oscillator(
fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['know_sure_thing'] = know_sure_thing(fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['average_true_range'] = average_true_range(
fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['adx'] = average_directional_index(
fund,
atr=fund_data['average_true_range']['tabular'],
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['parabolic_sar'] = parabolic_sar(
fund,
adx_tabular=fund_data['adx']['tabular'],
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
fund_data['demand_index'] = demand_index(fund,
plot_output=False,
name=fund_name,
progress_bar=p,
view=period)
if 'no_index' not in config['state']:
strength, match_data = relative_strength(
fund_name,
full_data_dict=dataset[period],
config=config,
plot_output=False,
meta=analysis[fund_name]['metadata'],
progress_bar=p,
period=period,
interval=config['interval'][i],
view=period)
fund_data['relative_strength'] = strength
fund_data['statistics']['risk_ratios'] = risk_comparison(
fund,
dataset[period]['^GSPC'],
dataset[period]['^IRX'],
sector_data=match_data)
p.uptick()
# Support and Resistance Analysis
fund_data['support_resistance'] = find_resistance_support_lines(
fund,
name=fund_name,
plot_output=False,
progress_bar=p,
view=period)
# Feature Detection Block
fund_data['features'] = {}
fund_data['features'][
'head_shoulders'] = feature_detection_head_and_shoulders(
fund,
name=fund_name,
plot_output=False,
progress_bar=p,
view=period)
fund_data['candlesticks'] = candlesticks(fund,
name=fund_name,
plot_output=False,
view=period,
progress_bar=p)
fund_data['price_gaps'] = analyze_price_gaps(fund,
name=fund_name,
plot_output=False,
progress_bar=p,
view=period)
# Get Trendlines
fund_data['trendlines'] = get_trendlines(
fund,
name=fund_name,
plot_output=False,
progress_bar=p,
view=period,
meta=analysis[fund_name]['metadata'])
# Various Fund-specific Metrics
fund_data['futures'] = future_returns(fund, progress_bar=p)
# Parse through indicators and pull out latest signals (must be last)
fund_data['last_signals'] = assemble_last_signals(
fund_data,
fund=fund,
name=fund_name,
view=period,
progress_bar=p,
plot_output=False)
p.end()
analysis[fund_name][period] = fund_data
analysis[fund_name]['synopsis'] = generate_synopsis(analysis,
name=fund_name)
return analysis, clock