We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
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
Hi,
In parameter file dqn_example.json#L17-L20 to train DQN model, the actions are
"actions": [ "4", "5" ],
What does it mean given cartpole-v0 has only two actions {0, 1}?
{0, 1}
Thanks
The text was updated successfully, but these errors were encountered:
@MisterTea Any info or suggestion will be very helpful. I tried to print details during evaluation and I still haven't figure it out
Float State Features = [{'0': 0.16316721, '1': 1.7206059, '2': -0.22225317, '3': -2.8040316}] Prediction: [{'4': -1.6168417, '5': 0.6317559}] Max Q values = 5 Action Index = 1
My guess so far is that actions values from dqn_example.json is mapped as {'4': 0, '5': 1}
actions
dqn_example.json
{'4': 0, '5': 1}
Sorry, something went wrong.
Hey! The actions are integers starting where the state integers leave off. In this case, the state features are [0...3]
The action feature IDs and state feature IDs can't overlap, which is why the actions are not 0 and 1
Thank you so much @MisterTea.
No branches or pull requests
Hi,
In parameter file dqn_example.json#L17-L20 to train DQN model, the actions are
What does it mean given cartpole-v0 has only two actions
{0, 1}
?Thanks
The text was updated successfully, but these errors were encountered: