Giters
RobertTLange
/
gymnax
RL Environments in JAX π
Geek Repo:
Geek Repo
Github PK Tool:
Github PK Tool
Stargazers:
535
Watchers:
10
Issues:
50
Forks:
48
RobertTLange/gymnax Issues
Differentiate step function ?
Updated
6 months ago
Comments count
6
GymnaxtoBrax Wrapper imports failing
Updated
8 months ago
`LogWrapper` should indicate missing data until first episode terminates
Updated
9 months ago
If `Environment.observation_space` requires the `EnvParams`, `Environment.get_obs` should too.
Updated
a year ago
On the use of `jnp.int_`
Updated
a year ago
Gymnax 0.0.6 requires gym>=0.26, but Visualizer asserts gym==0.19.0
Updated
a year ago
Comments count
1
Potential bug due to lax.select usage in step function
Closed
a year ago
Comments count
2
[Proposal] Environment API changes
Updated
a year ago
Comments count
3
Gymnasium API update
Updated
a year ago
Comments count
1
env step accumulates memory
Closed
a year ago
Comments count
1
How to use `env.render()` to visualize environment transitions frame by frame?
Updated
a year ago
Comments count
1
Potential bug in Tuple space
Closed
a year ago
Comments count
1
ValueError: mutable default <class 'jaxlib.xla_extension.ArrayImpl'> for field reward_timestep is not allowed: use default_factory
Closed
a year ago
Comments count
2
BernoulliBandit observation space bounds are incorrect when time normalisation is enabled.
Closed
a year ago
Comments count
1
Mention request for Pgx
Closed
a year ago
Pendulum-1, MountainCarContinuous-v0 and Reacher-misc return non-squeezed reward
Closed
a year ago
Modifying optimal return parameter has no effect (bug)
Closed
a year ago
Add A2C example notebook
Closed
a year ago
[Proposal] Gym conversion wrappers
Closed
a year ago
Comments count
4
Add Brax <-> Gymnax wrappers
Closed
a year ago
Issue: vmapped CartPole input shape does not match
Closed
2 years ago
Comments count
1
Add `Seaquest` MinAtar environment
Updated
2 years ago
AttributeError: module 'jax' has no attribute 'tree_multimap'
Closed
2 years ago
Comments count
1
Trained baselines values incl. active training
Closed
2 years ago
Comments count
1
Notebook links missing?
Closed
2 years ago
Comments count
2
CPU/GPU/TPU Benchmarks
Closed
2 years ago
Comments count
1
Action wrappers
Closed
2 years ago
Elegant registration of environments
Closed
2 years ago
MinAtar Environment Implementation
Closed
2 years ago
Automated tools for benchmarking
Closed
2 years ago
Comments count
1
Add TypeChecking to `gymnax`
Closed
2 years ago
`lax.select` versus `(1-x)*y + x*z`
Closed
2 years ago
Refactor `agents`/`dojos` into `experimental`
Closed
3 years ago
`TrajectoryCollector` with discount masking if terminal
Closed
3 years ago
Miscellaneous environments
Closed
3 years ago
Four Rooms (Sutton et al. 1999) environment
Closed
3 years ago
Add different requirement files
Closed
3 years ago
Comments count
1
DQN rlax + bsuite vs rlax + gymnax
Closed
3 years ago
Comments count
2
bsuite environment implementation
Closed
3 years ago
Comments count
1
Jittable `Environment` class
Closed
3 years ago
Comments count
1
Observation/Action Space Information & Sampling
Closed
3 years ago
Comments count
2
Add CONTRIBUTING .md file
Closed
3 years ago
Replace all `state` variables with dictionaries
Closed
3 years ago
Comments count
1
Replace all `params_env_name ` with `FrozenDict`
Closed
3 years ago
Comments count
1