def __init__(self, results):

self.fitness, self.ratio, self.models = results

def __str__(self):

def __init__(self):

self.digits, self.spread = load_symbol_properties()

self.db_indivs = load_indivs().to_numpy()

self.db_results = []

self.quotes = []

self.params = None

self.quotes_time = [np.resize(np.array(quotes).astype(int), 400) for quotes in load_quotes_time()]

for result in load_results():

self.quotes.append(np.array(result[6:]).astype(float))

self.db_results.append(np.array(result[:6]).astype(float))

self.db_results = np.array(self.db_results)

self.quotes = np.array(self.quotes)

self.quotes_time = np.array(self.quotes_time)

def calc(self, indivs: list, gpu, multithreaded, compile):

trading_params = []

nontrading_params = []

for indiv in indivs:

trading_params.append( np.array([gene.value() for gene in indiv.get("trading")]) )

nontrading_params.append( np.array([[gene.value() - gene.radius(), gene.value() + gene.radius()] for gene in indiv.get("trading", has_tag=False)]) )

results = []

if compile:

for indiv_idx in range(len(indivs)):

results.append(evaluate_compile(self.db_indivs, self.db_results, self.quotes, self.quotes_time, trading_params[indiv_idx], nontrading_params[indiv_idx], self.spread, self.digits))

elif multithreaded:

pool = Pool()

results = pool.map(evaluate, [self.db_indivs, self.db_results, self.quotes, self.quotes_time, trading_params[indiv_idx][indiv_idx], nontrading_params[indiv_idx], self.spread, self.digits])

pool.close()

pool.join()

elif gpu:

raise Exception("Gpu not implemented")

else:

for indiv_idx in range(len(indivs)):

results.append(evaluate([self.db_indivs, self.db_results, self.quotes, self.quotes_time, trading_params[indiv_idx], nontrading_params[indiv_idx], self.spread, self.digits]))

for indiv_idx, indiv in enumerate(indivs):

db_indivs, db_results, quotes, quotes_time, trading_params, non_trading_params, spread, digits = params

entry_dist, stop, stop_out, take, bu, bu_cond, min_dist_betw_orders, max_stops, exp, min_tim_betw_odrs = trading_params

trade_list = []

use_bu = True

if bu > bu_cond:

use_bu = False

# setting parameters

result = 0

result_list = []

orders = 0

trades = 0

models = 0

model_indexes = []

models_id = []

model_stops = []

model_takes = []

model_bus = []

max_drawdown = 0

max_account_value = 0

open_models = [] # list with OpenModel objects

recently_opened_models = [] # list of model numbers

for mod_idx in range(len(db_indivs)):

model = db_indivs[mod_idx]

model_number, model_time, level_time, level, atr, direction = db_results[mod_idx]

quote = quotes[mod_idx]

highs = quote[:int(len(quote) / 2)]

lows = quote[int(len(quote) / 2):]

time = quotes_time[mod_idx]

fits = True

open_ord_cond_shift = -spread if direction else 0 # (-spread) if buy, (0) otherwise

close_ord_cond_shift = 0 if direction else -spread # (0) if buy, (-spread) otherwise

placed_order = False

is_opened = False # order is in trade now

exp_time = math.inf

stops = 0

placed_bu = False

have_take = False

closed_by_bu = False

time_after_prev_order = math.inf # initiates with math.inf to pass (time_after_prev_order < min_tim_betw_odrs) condition and place first order

time_after_order = 0 # time passed after placing an order and before opening it

if direction: # long model

entry_lvl = round(level - entry_dist * atr, digits)

bu_lvl = round(entry_lvl + bu * atr, digits)

bu_condition_level = entry_lvl + bu_cond * atr

take_lvl = round(entry_lvl + take * atr, digits)

stop_lvl = round(entry_lvl - stop * atr, digits)

stop_out_lvl = entry_lvl - stop_out * atr

else: # short model

entry_lvl = round(level + entry_dist * atr, digits)

bu_lvl = round(entry_lvl - bu * atr, digits)

bu_condition_level = entry_lvl - bu_cond * atr

take_lvl = round(entry_lvl - take * atr, digits)

stop_lvl = round(entry_lvl + stop * atr, digits)

stop_out_lvl = entry_lvl + stop_out * atr

# bringing buffers up to date

for open_model in open_models:

if model_time >= open_model.close_time:

open_models.remove(open_model)

# checking if this model should be counted

for opened_model in recently_opened_models:

if model_number == opened_model:

fits = False

for open_model in open_models:

if open_model.range_start <= entry_lvl <= open_model.range_end:

fits = False

for gene_idx in range(len(model)):

attr = model[gene_idx]

if non_trading_params[gene_idx][0] > attr or non_trading_params[gene_idx][1] < attr:

fits = False

break

if not fits:

continue

# evaluating model

models += 1

price_is_up = False

if highs[0] > entry_lvl + open_ord_cond_shift and lows[0] > entry_lvl + open_ord_cond_shift:

price_is_up = True

current_time = time[0]

for idx in range(len(highs)):

current_time = time[idx]

time_after_prev_order += 1

if (direction and lows[idx] <= stop_out_lvl) or (not direction and highs[idx] >= stop_out_lvl):

if is_opened:

stops += 1

break

if is_opened:

if use_bu and not placed_bu and \

((direction and highs[idx] >= bu_condition_level) or (

not direction and lows[idx] <= bu_condition_level)):

stop_lvl = bu_lvl

placed_bu = True

elif (direction and lows[idx] <= stop_lvl + close_ord_cond_shift) or (

not direction and highs[idx] >= stop_lvl + close_ord_cond_shift):

if placed_bu:

closed_by_bu = True

break

stops += 1

if stops >= max_stops:

break

is_opened = False

time_after_order = 0

placed_order = False

if highs[idx] > entry_lvl + open_ord_cond_shift and lows[idx] > entry_lvl + open_ord_cond_shift:

price_is_up = True

else:

price_is_up = False

elif (direction and highs[idx] >= take_lvl + close_ord_cond_shift) or (

not direction and lows[idx] <= take_lvl + close_ord_cond_shift):

have_take = True

break

if not is_opened:

if not placed_order and time_after_prev_order < min_tim_betw_odrs: # if 'True' then we can't open new order right now

continue

if not placed_order:

time_after_prev_order = 0

placed_order = True

if current_time >= exp_time:

break

if (price_is_up and lows[idx] <= entry_lvl + open_ord_cond_shift) or \

(not price_is_up and highs[idx] >= entry_lvl + open_ord_cond_shift):

is_opened = True

else:

time_after_order += 1

result -= stops * stop

if have_take:

result += take

elif closed_by_bu:

result += bu

result = round(result, 2)

if result > max_account_value:

max_account_value = result

drawdown = round(max_account_value - result, 2)

if drawdown > max_drawdown:

max_drawdown = drawdown

open_models.append(Model(

entry_lvl - min_dist_betw_orders * atr,

entry_lvl + min_dist_betw_orders * atr,

current_time,

mod_idx + 1

))

if len(recently_opened_models) >= 10:

recently_opened_models.pop(0)

recently_opened_models.append(model_number)

ratio = result / max_drawdown if max_drawdown != 0 else 0

fitness = result

entry_dist, stop, stop_out, take, bu, bu_cond, min_dist_betw_orders, max_stops, exp, min_tim_betw_odrs = trading_params

#trade_list = []

use_bu = True

if bu > bu_cond:

use_bu = False

# setting parameters

result = 0

# result_list = []

# orders = 0

# trades = 0

models = 0

# model_indexes = []

# models_id = []

# model_stops = []

# model_takes = []

# model_bus = []

max_drawdown = 0

max_account_value = 0

open_models = [] # list with OpenModel objects

recently_opened_models = [] # list of model numbers

for mod_idx in range(len(db_indivs)):

model = db_indivs[mod_idx]

model_number, model_time, level_time, level, atr, direction = db_results[mod_idx]

quote = quotes[mod_idx]

highs = quote[:int(len(quote) / 2)]

lows = quote[int(len(quote) / 2):]

time = quotes_time[mod_idx]

fits = True

open_ord_cond_shift = -spread if direction else 0 # (-spread) if buy, (0) otherwise

close_ord_cond_shift = 0 if direction else -spread # (0) if buy, (-spread) otherwise

placed_order = False

is_opened = False # order is in trade now

exp_time = math.inf

stops = 0

placed_bu = False

have_take = False

closed_by_bu = False

time_after_prev_order = math.inf # initiates with math.inf to pass (time_after_prev_order < min_tim_betw_odrs) condition and place first order

time_after_order = 0 # time passed after placing an order and before opening it

if direction: # long model

entry_lvl = round(level - entry_dist * atr, digits)

bu_lvl = round(entry_lvl + bu * atr, digits)

bu_condition_level = entry_lvl + bu_cond * atr

take_lvl = round(entry_lvl + take * atr, digits)

stop_lvl = round(entry_lvl - stop * atr, digits)

stop_out_lvl = entry_lvl - stop_out * atr

else: # short model

entry_lvl = round(level + entry_dist * atr, digits)

bu_lvl = round(entry_lvl - bu * atr, digits)

bu_condition_level = entry_lvl - bu_cond * atr

take_lvl = round(entry_lvl - take * atr, digits)

stop_lvl = round(entry_lvl + stop * atr, digits)

stop_out_lvl = entry_lvl + stop_out * atr

# bringing buffers up to date

for open_model in open_models:

if model_time >= open_model.close_time:

open_models.remove(open_model)

# checking if this model should be counted

for opened_model in recently_opened_models:

if model_number == opened_model:

fits = False

for open_model in open_models:

if open_model.range_start <= entry_lvl <= open_model.range_end:

fits = False

for gene_idx in range(len(model)):

attr = model[gene_idx]

if non_trading_params[gene_idx][0] > attr or non_trading_params[gene_idx][1] < attr:

fits = False

break

if not fits:

continue

# evaluating model

models += 1

price_is_up = False

if highs[0] > entry_lvl + open_ord_cond_shift and lows[0] > entry_lvl + open_ord_cond_shift:

price_is_up = True

current_time = time[0]

for idx in range(len(highs)):

current_time = time[idx]

time_after_prev_order += 1

if (direction and lows[idx] <= stop_out_lvl) or (not direction and highs[idx] >= stop_out_lvl):

if is_opened:

stops += 1

break

if is_opened:

if use_bu and not placed_bu and \

((direction and highs[idx] >= bu_condition_level) or (

not direction and lows[idx] <= bu_condition_level)):

stop_lvl = bu_lvl

placed_bu = True

elif (direction and lows[idx] <= stop_lvl + close_ord_cond_shift) or (

not direction and highs[idx] >= stop_lvl + close_ord_cond_shift):

if placed_bu:

closed_by_bu = True

break

stops += 1

if stops >= max_stops:

break

is_opened = False

time_after_order = 0

placed_order = False

if highs[idx] > entry_lvl + open_ord_cond_shift and lows[idx] > entry_lvl + open_ord_cond_shift:

price_is_up = True

else:

price_is_up = False

elif (direction and highs[idx] >= take_lvl + close_ord_cond_shift) or (

not direction and lows[idx] <= take_lvl + close_ord_cond_shift):

have_take = True

break

if not is_opened:

if not placed_order and time_after_prev_order < min_tim_betw_odrs: # if 'True' then we can't open new order right now

continue

if not placed_order:

time_after_prev_order = 0

placed_order = True

if current_time >= exp_time:

break

if (price_is_up and lows[idx] <= entry_lvl + open_ord_cond_shift) or \

(not price_is_up and highs[idx] >= entry_lvl + open_ord_cond_shift):

is_opened = True

else:

time_after_order += 1

result -= stops * stop

if have_take:

result += take

elif closed_by_bu:

result += bu

result = round(result, 2)

if result > max_account_value:

max_account_value = result

drawdown = round(max_account_value - result, 2)

if drawdown > max_drawdown:

max_drawdown = drawdown

open_models.append(Model(

entry_lvl - min_dist_betw_orders * atr,

entry_lvl + min_dist_betw_orders * atr,

current_time,

mod_idx + 1

))

if len(recently_opened_models) >= 10:

recently_opened_models.pop(0)

recently_opened_models.append(model_number)

ratio = result / max_drawdown if max_drawdown != 0 else 0

fitness = result