Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Devs/restruc #5

Open
wants to merge 7 commits into
base: master
Choose a base branch
from
Open

Devs/restruc #5

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
a25cc3b
support matplotlib show=false render and still plot the fig
Yvictor Dec 7, 2017
7d75cc1
customize return transaction into state
Yvictor Dec 10, 2017
b4c3739
fixed return transaction features len for build model shape with retu…
Yvictor Dec 16, 2017
381f838
add step with custom done support for set loss max
Yvictor Dec 19, 2017
94e0a5a
fixed return transaction when call step not updated
Yvictor Apr 5, 2018
6d2bd10
v2 init
Yvictor Apr 5, 2018
dc2e173
add action 3 for cover, modify action long and short include cover an…
Yvictor Apr 6, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
25 changes: 25 additions & 0 deletions trading_env/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,31 @@ def available_envs():
def make(env_id, obs_data_len, step_len,
df, fee, deal_col_name='price',
feature_names=['price', 'volume'],
return_transaction=False,
*args , **kwargs):
"""
v0: currently not maintain

v1:
# assert df
# need deal price as essential and specified the df format
# obs_data_leng -> observation data length
# step_len -> when call step rolling windows will + step_len
# df -> dataframe that contain data for trading(format as...)
# price
# datetime
# serial_number -> serial num of deal at each day recalculating

# fee : when each deal will pay the fee, set with your product
# max_position : the max market position for you trading share
# deal_col_name : the column name for cucalate reward used.
# feature_names : list contain the feature columns to use in trading status.
# return_transaction : bool, list or dict default False
True will return all column include 'mkt_pos', 'mkt_pos_var', 'entry_cover', 'avg_hold_price', 'fluc_reward', 'make_real', 'reward'
use list to define which column to return
use dict to define which column to return and dict value as the func apply to array
# ?day trade option set as default if don't use this need modify
"""
envs = available_envs()
assert env_id in envs , "env_id: {} not exist. try one of {}".format(env_id, str(envs).strip('[]'))
assert deal_col_name in df.columns, "deal_col not in Dataframe please define the correct column name of which column want to calculate the profit."
Expand All @@ -22,5 +46,6 @@ def make(env_id, obs_data_len, step_len,
step_len=step_len ,df=df, fee=fee,
deal_col_name=deal_col_name,
feature_names=feature_names,
return_transaction=return_transaction,
*args, **kwargs)
return env
4 changes: 3 additions & 1 deletion trading_env/envs/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,8 @@
from . import training_v0
from . import training_v1
from . import training_v2
from . import backtest_v0
from . import backtest_v1
from . import backtest_v2

available_envs_module = [training_v0, training_v1, backtest_v0, backtest_v1]
available_envs_module = [training_v0, training_v1, training_v2, backtest_v0, backtest_v1, backtest_v2]
100 changes: 73 additions & 27 deletions trading_env/envs/backtest_v1.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ class trading_env:
def __init__(self, env_id, obs_data_len, step_len,
df, fee, max_position=5, deal_col_name='price',
feature_names=['price', 'volume'],
return_transaction=True,
return_transaction=False,
fluc_div=100.0, gameover_limit=5,
*args, **kwargs):
"""
Expand Down Expand Up @@ -42,9 +42,18 @@ def __init__(self, env_id, obs_data_len, step_len,
self.action_describe = {0:'do nothing',
1:'long',
2:'short'}



if isinstance(return_transaction, bool) and return_transaction:
transaction_num = 7
elif not return_transaction:
transaction_num = 0
else:
transaction_num = len(return_transaction)

self.obs_len = obs_data_len
self.feature_len = len(feature_names)
self.return_state_len = len(feature_names) + transaction_num
self.observation_space = np.array([self.obs_len*self.feature_len,])
self.using_feature = feature_names
self.price_name = deal_col_name
Expand Down Expand Up @@ -117,17 +126,35 @@ def reset(self):
self.obs_reward_fluctuant = self.reward_fluctuant_arr[self.step_st: self.step_st+self.obs_len]
self.obs_makereal = self.reward_makereal_arr[self.step_st: self.step_st+self.obs_len]
self.obs_reward = self.reward_arr[self.step_st: self.step_st+self.obs_len]

self.transaction_dict = {'mkt_pos': self.obs_posi[:, np.newaxis],
'mkt_pos_var': self.obs_posi_var[:, np.newaxis],
'entry_cover': self.obs_posi_entry_cover[:, np.newaxis],
'avg_hold_price': self.obs_price_mean[:, np.newaxis],
'fluc_reward': self.obs_reward_fluctuant[:, np.newaxis],
'make_real': self.obs_makereal[:, np.newaxis],
'reward': self.obs_reward[:, np.newaxis],}

if self.return_transaction:
self.obs_return = np.concatenate((self.obs_state,
self.obs_posi[:, np.newaxis],
self.obs_posi_var[:, np.newaxis],
self.obs_posi_entry_cover[:, np.newaxis],
self.obs_price[:, np.newaxis],
self.obs_price_mean[:, np.newaxis],
self.obs_reward_fluctuant[:, np.newaxis],
self.obs_makereal[:, np.newaxis],
self.obs_reward[:, np.newaxis]), axis=1)
if isinstance(self.return_transaction, bool):
self.obs_return = np.concatenate((self.obs_state,
self.obs_posi[:, np.newaxis],
self.obs_posi_var[:, np.newaxis],
self.obs_posi_entry_cover[:, np.newaxis],
self.obs_price_mean[:, np.newaxis],
self.obs_reward_fluctuant[:, np.newaxis],
self.obs_makereal[:, np.newaxis],
self.obs_reward[:, np.newaxis]), axis=1)
elif isinstance(self.return_transaction, list):
self.obs_return = np.concatenate((self.obs_state,)+ \
tuple(self.transaction_dict[need] \
for need in self.return_transaction),
axis=1)
elif isinstance(self.return_transaction, dict):
self.obs_return = np.concatenate((self.obs_state,)+ \
tuple(self.return_transaction[need](self.transaction_dict[need]) \
for need in self.return_transaction),
axis=1)
else:
self.obs_return = self.obs_state

Expand Down Expand Up @@ -183,7 +210,7 @@ def _stayon(self, current_price_mean, current_mkt_position):
self.chg_posi[:] = current_mkt_position
self.chg_price_mean[:] = current_price_mean

def step(self, action):
def step(self, action, custom_done=False):
current_index = self.step_st + self.obs_len -1
current_price_mean = self.price_mean_arr[current_index]
current_mkt_position = self.posi_arr[current_index]
Expand Down Expand Up @@ -213,7 +240,7 @@ def step(self, action):
self.chg_reward = self.obs_reward[-self.step_len:]

done = False
if self.step_st+self.obs_len+self.step_len >= len(self.price):
if self.step_st+self.obs_len+self.step_len >= len(self.price) or custom_done:
done = True
action = -1
if current_mkt_position != 0:
Expand Down Expand Up @@ -265,15 +292,32 @@ def step(self, action):
self.chg_reward_fluctuant[:] = (self.chg_price - self.chg_price_mean)*self.chg_posi - np.abs(self.chg_posi)*self.fee

if self.return_transaction:
self.obs_return = np.concatenate((self.obs_state,
self.obs_posi[:, np.newaxis],
self.obs_posi_var[:, np.newaxis],
self.obs_posi_entry_cover[:, np.newaxis],
self.obs_price[:, np.newaxis],
self.obs_price_mean[:, np.newaxis],
self.obs_reward_fluctuant[:, np.newaxis],
self.obs_makereal[:, np.newaxis],
self.obs_reward[:, np.newaxis]), axis=1)
self.transaction_dict = {'mkt_pos': self.obs_posi[:, np.newaxis],
'mkt_pos_var': self.obs_posi_var[:, np.newaxis],
'entry_cover': self.obs_posi_entry_cover[:, np.newaxis],
'avg_hold_price': self.obs_price_mean[:, np.newaxis],
'fluc_reward': self.obs_reward_fluctuant[:, np.newaxis],
'make_real': self.obs_makereal[:, np.newaxis],
'reward': self.obs_reward[:, np.newaxis],}
if isinstance(self.return_transaction, bool):
self.obs_return = np.concatenate((self.obs_state,
self.obs_posi[:, np.newaxis],
self.obs_posi_var[:, np.newaxis],
self.obs_posi_entry_cover[:, np.newaxis],
self.obs_price_mean[:, np.newaxis],
self.obs_reward_fluctuant[:, np.newaxis],
self.obs_makereal[:, np.newaxis],
self.obs_reward[:, np.newaxis]), axis=1)
elif isinstance(self.return_transaction, list):
self.obs_return = np.concatenate((self.obs_state,)+ \
tuple(self.transaction_dict[need] \
for need in self.return_transaction),
axis=1)
elif isinstance(self.return_transaction, dict):
self.obs_return = np.concatenate((self.obs_state,)+ \
tuple(self.return_transaction[need](self.transaction_dict[need]) \
for need in self.return_transaction),
axis=1)
else:
self.obs_return = self.obs_state

Expand Down Expand Up @@ -339,7 +383,7 @@ def _plot_trading(self):
c=trade_color_sell, edgecolors=(0,1,0,0.9), zorder=2)


def render(self, save=False):
def render(self, save=False, show=True):
if self.render_on == 0:
matplotlib.style.use('dark_background')
self.render_on = 1
Expand All @@ -363,9 +407,10 @@ def render(self, save=False):
self._plot_trading()

self.ax.set_xlim(0,len(self.price[:self.step_st+self.obs_len])+200)
plt.ion()
#self.fig.tight_layout()
plt.show()
if show:
plt.ion()
#self.fig.tight_layout()
plt.show()
if save:
self.fig.savefig('fig/%s.png' % str(self.t_index))

Expand All @@ -387,4 +432,5 @@ def render(self, save=False):
self.ax.set_xlim(0,len(self.price[:self.step_st+self.obs_len])+200)
if save:
self.fig.savefig('fig/%s.png' % str(self.t_index))
plt.pause(0.0001)
if show:
plt.pause(0.0001)
Loading