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

Jump to bottom

SAC and cross-maze ant #8

Open
acohen13 opened this issue Jun 12, 2018 · 2 comments
Open

SAC and cross-maze ant #8

acohen13 opened this issue Jun 12, 2018 · 2 comments

Comments

@acohen13
Copy link

Hi,

Can SAC run with the cross-maze variation for ant? With the default parameters, the command:

"python ./examples/mujoco_all_sac.py --env=ant --domain=ant --task=cross-maze --policy=gmm --log_dir=data/ant_cross-experiment"

does not throw any errors but "Launches 0 experiments." I was able to track the problem down to rllab's _ivariants_sorted function in class VariantGenerator in misc/instrument.py which returns a generator for an empty list when the cross-maze task is specified (it seems to work for Multidirection) as a opposed to a list with a dictionary containing run parameters.

Am I doing something wrong or not including a run flag?

Other than that, thanks for sharing the code!
hartikainen pushed a commit that referenced this issue Jun 20, 2018
@haarnoja @hartikainen
Real nvp/policy regularization (#8)
6284537 
* Apply nonlinearity to preprocessor output
* Regularize scale directly instead of the scale network parameters
* Add scale regularization losses into 'tf.GraphKeys.REGULARIATION_LOSSES'
@ahq1993
Copy link

ahq1993 commented Nov 23, 2018

Hi,
Thanks for uploading the code.
I am running into the same issue. Not able to run ant: cross-maze. It throws "Launches 0 experiments" and does nothing.

hartikainen pushed a commit that referenced this issue Feb 24, 2019
@haarnoja
Real nvp/policy regularization (#8)
82614ce 
* Apply nonlinearity to preprocessor output
* Regularize scale directly instead of the scale network parameters
* Add scale regularization losses into 'tf.GraphKeys.REGULARIATION_LOSSES'
@GbengaOdesanmi
Copy link

Hi, thanks for uploading this code.
Please I am having this error while trying to run the code

OSError: /mujoco_models/libmujoco150.so: cannot open shared object file: No such file or directory

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@acohen13 @ahq1993 @GbengaOdesanmi