OpenAI Gym is a platform for reinforcement learning research that aims to provide a general-intelligence benchmark with a wide variety of environments.
Questions tagged [openai-gym]
1033 questions
-1
votes
1 answer
Convolution for state representation
When using DQN, other deep RL algorithms, does it make sense to use convolutional layer in the actor or critic network when you have a state input?
Let's say:
state representation 1: (obj label, position, velocity) of each object in the…
user3180
- 1,369
- 1
- 21
- 38
-1
votes
1 answer
Deep Q Network not Solving OpenAI CartPole
I was trying to implement a DQN to solve the CartPole-v0 task in the OpenAI Gym. Unfortunately, my implementation's performance does not seem to be improving.
Currently, as the training occurs, the episode reward actually decreases whereas the goal…
ashboy64
- 83
- 3
- 13
-1
votes
1 answer
Why do openai gym return reward zero for terminal states?
I've been experimenting with Gym (and RL) a lot lately and there is one specific behaviour of gym that has piqued my interest. Why is it that OpenAI Gym return reward 0 even when game is over? For e.g, in Breakout-v0, when all five lives are spent,…
Nilesh PS
- 356
- 3
- 8
-1
votes
1 answer
AttributeError: 'NoneType' object has no attribute 'flip' when using gym
I tried the code in open AI gym documentation. It works but errors occur after I close the window. The code I used is below:
import gym
env = gym.make('CartPole-v0')
for i_episode in range(20):
observation = env.reset()
for t in range(100):
…
D1cvv0ng
- 41
- 3
-1
votes
1 answer
Can't install openai universe with pip install -e?
I get this error while I follow the install rule that provides the company. Here is the install [link]: (https://github.com/openai/universe#installation)
showing this error:
user@user:~/universe$ pip install -e
Usage:
pip install [options]…
RajuHossen
- 1
- 1
-2
votes
0 answers
Problem with training Deep Q-learning model on local machine
I am attempting to solve an OpenAI Gym environment using Deep Q Learning with TensorFlow on an M1 Mac. The model seems to start fitting well but ends up performing worse every time. In contrast, when I used the same code in Google Colab, it worked…
Sid
- 1
-2
votes
1 answer
Reinforcement learning agent picking same action for every state - OpenAI gym environment with stable_baselines3
I have been training a reinforcement learning agent to play ultimate-tictactoe (an expanded version of tic-tac-toe with a 9x9 board and additional rules).
I've created an openai gym environment, and have been trying to train the agent using…
Samatva K
- 1
- 1
-2
votes
1 answer
NameError: name 'env' is not defined
I am trying to train a model to solve the FrozenLake-v0 problem.
In the process am trying to instantiate the environment in the following way. But encountering an error. Please help me with this
!apt install xvfb -y
!pip install…
Vasudha Pasumarthi
- 1
- 1
- 3
-2
votes
2 answers
pip install is working in notebook cell but not working in script
I am trying pip commands. I am installing OpenAI Gym. I wrote the following command in a notebook cell and run it: pip install gym It worked and gave me a message of successful installation.
But when I tryed typing the same command in a .py script…
Osama El-Ghonimy
- 335
- 2
- 12
-2
votes
1 answer
__init__() missing 1 required positional argument 'master'
I'm having an issue with creating this main file with a custom environment for open-AI. So I have created the environment and everything installs OK but when i try to run it it shows the error in the title.
Here is the Main file.
class Params():
…
Marc Leese
- 31
- 1
- 6
-2
votes
1 answer
How to render from ob openCV
In my python neat code I am using opencv to downscale and covert into gray every frame of the environment. what I want a archive is that opencv opens a window displaying the frame/ video that is it processing.
In short I want to view the the neat…
Jim Hoggey
- 9
- 5
-3
votes
1 answer
About testing Deep Q-Network after training.Training data and test data do not correspond
I am using deep q network breakout to play Atari Breakout.
Some of the latest training results：
running reward: 10.19 at episode 19285, frame count 1900000
running reward: 9.95 at episode 19320, frame count 1910000
running reward: 9.12 at episode…
John
- 1
-4
votes
1 answer
Errors with 'pip' and 'install'
I have python 3.5 installed on OSX, however 'pip' does not work for some reason and 'install' is a syntax error, could some body give me step by step beginners directions to get this sorted.
AI-Jay
- 1