Merge pull request #472 from tensortrade-org/update-dependencies

Bump to gymnasium and update dependencies to Python 3.11.9

.travis.yml

@@ -6,7 +6,7 @@ services:
   - docker
 
 python:
-  - '3.9'
+  - '3.11.9'
 
 install:
   - sudo apt-get update

requirements.txt

@@ -1,6 +1,6 @@
 numpy>=1.17.0
 pandas>=0.25.0
-gym>=0.25.2
+gymnasium>=0.28.1
 pyyaml>=5.1.2
 stochastic>=0.6.0
 tensorflow>=2.7.0

setup.py

@@ -44,11 +44,11 @@
         if package.startswith('tensortrade')
     ],
     license='Apache 2.0',
-    python_requires='>=3.7',
+    python_requires='>=3.11.9',
     install_requires=[
         'numpy>=1.17.0',
         'pandas>=0.25.0',
-        'gym>=0.15.7',
+        'gymnasium>=0.28.1',
         'pyyaml>=5.1.2',
         'stochastic>=0.6.0',
         'tensorflow>=2.7.0',
@@ -83,9 +83,7 @@
         'Intended Audience :: Information Technology',
         'License :: OSI Approved :: Apache Software License',
         'Operating System :: OS Independent',
-        'Programming Language :: Python :: 3.7',
-        'Programming Language :: Python :: 3.8',
-        'Programming Language :: Python :: 3.9',
+        'Programming Language :: Python :: 3.11',
         'Topic :: Office/Business :: Financial :: Investment',
         'Topic :: Office/Business :: Financial',
         'Topic :: Scientific/Engineering :: Information Analysis',

tensortrade/env/default/actions.py

@@ -3,7 +3,7 @@
 from itertools import product
 from typing import Union, List, Any
 
-from gym.spaces import Space, Discrete
+from gymnasium.spaces import Space, Discrete
 
 from tensortrade.core import Clock
 from tensortrade.env.generic import ActionScheme, TradingEnv

tensortrade/env/default/observers.py

@@ -6,7 +6,7 @@
 import numpy as np
 import pandas as pd
 
-from gym.spaces import Box, Space
+from gymnasium.spaces import Box, Space
 from random import randrange
 
 

tensortrade/env/generic/components/action_scheme.py

@@ -15,7 +15,7 @@
 from abc import abstractmethod, ABCMeta
 from typing import Any
 
-from gym.spaces import Space
+from gymnasium.spaces import Space
 
 
 from tensortrade.core.component import Component

tensortrade/env/generic/components/observer.py

@@ -18,7 +18,7 @@
 
 import numpy as np
 
-from gym.spaces import Space
+from gymnasium.spaces import Space
 
 
 from tensortrade.core.component import Component

tensortrade/env/generic/environment.py

@@ -18,7 +18,7 @@
 from typing import Dict, Any, Tuple
 from random import randint
 
-import gym
+import gymnasium
 import numpy as np
 
 from tensortrade.core import TimeIndexed, Clock, Component
@@ -32,7 +32,7 @@
 )
 
 
-class TradingEnv(gym.Env, TimeIndexed):
+class TradingEnv(gymnasium.Env, TimeIndexed):
     """A trading environment made for use with Gym-compatible reinforcement
     learning algorithms.
 
@@ -81,13 +81,6 @@ def __init__(self,
         self.min_periods = min_periods
         self.random_start_pct = random_start_pct
 
-        # Register the environment in Gym and fetch spec
-        gym.envs.register(
-            id='TensorTrade-v0',
-            max_episode_steps=max_episode_steps,
-        )
-        self.spec = gym.spec(env_id='TensorTrade-v0')
-
         for c in self.components.values():
             c.clock = self.clock
 
@@ -135,14 +128,15 @@ def step(self, action: Any) -> 'Tuple[np.array, float, bool, dict]':
 
         obs = self.observer.observe(self)
         reward = self.reward_scheme.reward(self)
-        done = self.stopper.stop(self)
+        terminated = self.stopper.stop(self)
+        truncated = False
         info = self.informer.info(self)
 
         self.clock.increment()
 
-        return obs, reward, done, info
+        return obs, reward, terminated, truncated, info
 
-    def reset(self) -> 'np.array':
+    def reset(self,seed = None, options = None) -> tuple["np.array", dict[str, Any]]:
         """Resets the environment.
 
         Returns
@@ -167,10 +161,11 @@ def reset(self) -> 'np.array':
                     c.reset()
 
         obs = self.observer.observe(self)
+        info = self.informer.info(self)
 
         self.clock.increment()
 
-        return obs
+        return obs, info
 
     def render(self, **kwargs) -> None:
         """Renders the environment."""