3645506801

Build Type

push

github-actions

Committed by GitHub

Commit Message

Merge pull request #7818 from freqtrade/dependabot/pip/develop/torch-1.13.0

Run Details

16503 of 17351 relevant lines covered (95.11%)

0.95 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

91.67

/freqtrade/freqai/RL/Base4ActionRLEnv.py

import logging
from enum import Enum

from gym import spaces

from freqtrade.freqai.RL.BaseEnvironment import BaseEnvironment, Positions


logger = logging.getLogger(__name__)


class Actions(Enum):
    Neutral = 0
    Exit = 1
    Long_enter = 2
    Short_enter = 3


class Base4ActionRLEnv(BaseEnvironment):
    """
    Base class for a 4 action environment
    """

    def set_action_space(self):
        self.action_space = spaces.Discrete(len(Actions))

    def step(self, action: int):
        """
        Logic for a single step (incrementing one candle in time)
        by the agent
        :param: action: int = the action type that the agent plans
            to take for the current step.
        :returns:
            observation = current state of environment
            step_reward = the reward from `calculate_reward()`
            _done = if the agent "died" or if the candles finished
            info = dict passed back to openai gym lib
        """
        self._done = False
        self._current_tick += 1

        if self._current_tick == self._end_tick:
            self._done = True

        self._update_unrealized_total_profit()

        step_reward = self.calculate_reward(action)
        self.total_reward += step_reward

        trade_type = None
        if self.is_tradesignal(action):
            """
            Action: Neutral, position: Long ->  Close Long
            Action: Neutral, position: Short -> Close Short

            Action: Long, position: Neutral -> Open Long
            Action: Long, position: Short -> Close Short and Open Long

            Action: Short, position: Neutral -> Open Short
            Action: Short, position: Long -> Close Long and Open Short
            """

            if action == Actions.Neutral.value:
                self._position = Positions.Neutral
                trade_type = "neutral"
                self._last_trade_tick = None
            elif action == Actions.Long_enter.value:
                self._position = Positions.Long
                trade_type = "long"
                self._last_trade_tick = self._current_tick
            elif action == Actions.Short_enter.value:
                self._position = Positions.Short
                trade_type = "short"
                self._last_trade_tick = self._current_tick
            elif action == Actions.Exit.value:
                self._update_total_profit()
                self._position = Positions.Neutral
                trade_type = "neutral"
                self._last_trade_tick = None
            else:
                print("case not defined")

            if trade_type is not None:
                self.trade_history.append(
                    {'price': self.current_price(), 'index': self._current_tick,
                     'type': trade_type})

        if self._total_profit < 1 - self.rl_config.get('max_training_drawdown_pct', 0.8):
            self._done = True

        self._position_history.append(self._position)

        info = dict(
            tick=self._current_tick,
            total_reward=self.total_reward,
            total_profit=self._total_profit,
            position=self._position.value
        )

        observation = self._get_observation()

        self._update_history(info)

        return observation, step_reward, self._done, info

    def is_tradesignal(self, action: int) -> bool:
        """
        Determine if the signal is a trade signal
        e.g.: agent wants a Actions.Long_exit while it is in a Positions.short
        """
        return not ((action == Actions.Neutral.value and self._position == Positions.Neutral) or
                    (action == Actions.Neutral.value and self._position == Positions.Short) or
                    (action == Actions.Neutral.value and self._position == Positions.Long) or
                    (action == Actions.Short_enter.value and self._position == Positions.Short) or
                    (action == Actions.Short_enter.value and self._position == Positions.Long) or
                    (action == Actions.Exit.value and self._position == Positions.Neutral) or
                    (action == Actions.Long_enter.value and self._position == Positions.Long) or
                    (action == Actions.Long_enter.value and self._position == Positions.Short))

    def _is_valid(self, action: int) -> bool:
        """
        Determine if the signal is valid.
        e.g.: agent wants a Actions.Long_exit while it is in a Positions.short
        """
        # Agent should only try to exit if it is in position
        if action == Actions.Exit.value:
            if self._position not in (Positions.Short, Positions.Long):
                return False

        # Agent should only try to enter if it is not in position
        if action in (Actions.Short_enter.value, Actions.Long_enter.value):
            if self._position != Positions.Neutral:
                return False

        return True

1	import logging	1✔
2	from enum import Enum	1✔
3
4	from gym import spaces	1✔
5
6	from freqtrade.freqai.RL.BaseEnvironment import BaseEnvironment, Positions	1✔
7
8
9	logger = logging.getLogger(__name__)	1✔
10
11
12	class Actions(Enum):	1✔
13	Neutral = 0	1✔
14	Exit = 1	1✔
15	Long_enter = 2	1✔
16	Short_enter = 3	1✔
17
18
19	class Base4ActionRLEnv(BaseEnvironment):	1✔
20	"""
21	Base class for a 4 action environment
22	"""
23
24	def set_action_space(self):	1✔
25	self.action_space = spaces.Discrete(len(Actions))	1✔
26
27	def step(self, action: int):	1✔
28	"""
29	Logic for a single step (incrementing one candle in time)
30	by the agent
31	:param: action: int = the action type that the agent plans
32	to take for the current step.
33	:returns:
34	observation = current state of environment
35	step_reward = the reward from `calculate_reward()`
36	_done = if the agent "died" or if the candles finished
37	info = dict passed back to openai gym lib
38	"""
39	self._done = False	1✔
40	self._current_tick += 1	1✔
41
42	if self._current_tick == self._end_tick:	1✔
43	self._done = True	1✔
44
45	self._update_unrealized_total_profit()	1✔
46
47	step_reward = self.calculate_reward(action)	1✔
48	self.total_reward += step_reward	1✔
49
50	trade_type = None	1✔
51	if self.is_tradesignal(action):	1✔
52	"""
53	Action: Neutral, position: Long -> Close Long
54	Action: Neutral, position: Short -> Close Short
55
56	Action: Long, position: Neutral -> Open Long
57	Action: Long, position: Short -> Close Short and Open Long
58
59	Action: Short, position: Neutral -> Open Short
60	Action: Short, position: Long -> Close Long and Open Short
61	"""
62
63	if action == Actions.Neutral.value:	1✔
64	self._position = Positions.Neutral	×
65	trade_type = "neutral"	×
66	self._last_trade_tick = None	×
67	elif action == Actions.Long_enter.value:	1✔
68	self._position = Positions.Long	1✔
69	trade_type = "long"	1✔
70	self._last_trade_tick = self._current_tick	1✔
71	elif action == Actions.Short_enter.value:	1✔
72	self._position = Positions.Short	1✔
73	trade_type = "short"	1✔
74	self._last_trade_tick = self._current_tick	1✔
75	elif action == Actions.Exit.value:	1✔
76	self._update_total_profit()	1✔
77	self._position = Positions.Neutral	1✔
78	trade_type = "neutral"	1✔
79	self._last_trade_tick = None	1✔
80	else:
81	print("case not defined")	×
82
83	if trade_type is not None:	1✔
84	self.trade_history.append(	1✔
85	{'price': self.current_price(), 'index': self._current_tick,
86	'type': trade_type})
87
88	if self._total_profit < 1 - self.rl_config.get('max_training_drawdown_pct', 0.8):	1✔
89	self._done = True	×
90
91	self._position_history.append(self._position)	1✔
92
93	info = dict(	1✔
94	tick=self._current_tick,
95	total_reward=self.total_reward,
96	total_profit=self._total_profit,
97	position=self._position.value
98	)
99
100	observation = self._get_observation()	1✔
101
102	self._update_history(info)	1✔
103
104	return observation, step_reward, self._done, info	1✔
105
106	def is_tradesignal(self, action: int) -> bool:	1✔
107	"""
108	Determine if the signal is a trade signal
109	e.g.: agent wants a Actions.Long_exit while it is in a Positions.short
110	"""
111	return not ((action == Actions.Neutral.value and self._position == Positions.Neutral) or	1✔
112	(action == Actions.Neutral.value and self._position == Positions.Short) or
113	(action == Actions.Neutral.value and self._position == Positions.Long) or
114	(action == Actions.Short_enter.value and self._position == Positions.Short) or
115	(action == Actions.Short_enter.value and self._position == Positions.Long) or
116	(action == Actions.Exit.value and self._position == Positions.Neutral) or
117	(action == Actions.Long_enter.value and self._position == Positions.Long) or
118	(action == Actions.Long_enter.value and self._position == Positions.Short))
119
120	def _is_valid(self, action: int) -> bool:	1✔
121	"""
122	Determine if the signal is valid.
123	e.g.: agent wants a Actions.Long_exit while it is in a Positions.short
124	"""
125	# Agent should only try to exit if it is in position
126	if action == Actions.Exit.value:	1✔
127	if self._position not in (Positions.Short, Positions.Long):	1✔
128	return False	1✔
129
130	# Agent should only try to enter if it is not in position
131	if action in (Actions.Short_enter.value, Actions.Long_enter.value):	1✔
132	if self._position != Positions.Neutral:	1✔
133	return False	1✔
134
135	return True	1✔

freqtrade / freqtrade / 3645506801

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous