Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -69,15 +69,15 @@ The reward is the coverage of the block in the goal zone. The reward is 1.0 if t

### Success Criteria

The environment is considered solved if the block is at least 95% in the goal zone.
By default, the environment is considered solved if the block is at least 95% in the goal zone. The percent can be configured via `success_threshold` parameter.

### Starting State

The agent starts at a random position and the block starts at a random position and angle.

### Episode Termination

The episode terminates when the block is at least 95% in the goal zone.
The episode terminates when the success criteria is met.

### Arguments

Expand Down
9 changes: 6 additions & 3 deletions gym_pusht/envs/pusht.py
Original file line number Diff line number Diff line change
Expand Up @@ -70,15 +70,15 @@ class PushTEnv(gym.Env):

## Success Criteria

The environment is considered solved if the block is at least 95% in the goal zone.
By default, the environment is considered solved if the block is at least 95% in the goal zone. The percent can be configured via `success_threshold` parameter.

## Starting State

The agent starts at a random position and the block starts at a random position and angle.

## Episode Termination

The episode terminates when the block is at least 95% in the goal zone.
The episode terminates when the success criteria is met.

## Arguments

Expand All @@ -105,6 +105,8 @@ class PushTEnv(gym.Env):

* `visualization_height`: (int) The height of the visualized image. Default is `680`.

* `success_threshold`: (float) The percentage of block in goal zone required for environment to be considered solved. Default is `0.95`.

## Reset Arguments

Passing the option `options["reset_to_state"]` will reset the environment to a specific state.
Expand Down Expand Up @@ -144,6 +146,7 @@ def __init__(
observation_height=96,
visualization_width=680,
visualization_height=680,
success_threshold=0.95,
):
super().__init__()
# Observations
Expand Down Expand Up @@ -178,7 +181,7 @@ def __init__(
self.teleop = None
self._last_action = None

self.success_threshold = 0.95 # 95% coverage
self.success_threshold = success_threshold

def _initialize_observation_space(self):
if self.obs_type == "state":
Expand Down