Comments (4)
The two code snippets seem like different environments. Looks like your {env_id}
is not OpenCabinetDrawer-v1
. Note that OpenCabinetDrawer-v1
doesn't support pd_ee_delta_pose
(since the environment has a mobile base, you should use base_pd_joint_vel_arm_pd_joint_vel
as the controller, which is also the controller in demonstrations).
from maniskill.
Hi,
Thanks for your reply. I'm trying to match the environment up right now.
env_id = "OpenCabinetDrawer-v1"
!python -m mani_skill2.utils.download_asset partnet_mobility_cabinet
# download the raw dataset
!python -m mani_skill2.utils.download_demo {env_id} -o demos
# Directly download the converted demonstrations dataset files
!gdown "https://drive.google.com/drive/folders/1oNLgMVkByoHMv6mio-KxTkd7DrE6amKZ?usp=share_link" --folder -O "demos/rigid_body/{env_id}"
Retrieving folder list
Processing file 1QkkKDKfoJgu2Mv02687vFMEsGiJRkgx3 trajectory.h5
Processing file 1LfZPUN88izYPl7sjd6KK7FZSQnE8zleo trajectory.json
Processing file 1zrO6YDCOym8XfJbsg43H6VXkwqNpF9_Z trajectory.rgbd.pd_ee_delta_pose.h5
Processing file 1nInnP1uw9dyUuKdCwl_mtzAnnlbEoyso trajectory.rgbd.pd_ee_delta_pose.json
Processing file 1RSvvATA1UtPfA6lMJvdBXS4UY2Nn10-v trajectory.state.pd_ee_delta_pose.h5
Processing file 1kUxQfGp-gycODCxRnV6uulpnyxmIYHdJ trajectory.state.pd_ee_delta_pose.json
As you suggested, I used the trajectory.h5 file because everything else had the pd_ee_delta_pose on it.
# convert the dataset to include state observations.
!python -m mani_skill2.trajectory.replay_trajectory --traj-path \
demos/rigid_body/{env_id}/trajectory.h5 --save-traj \
--obs-mode state --target-control-mode pd_joint_vel --num-procs 10
dataset = ManiSkill2Dataset(f"demos/rigid_body/{env_id}/trajectory.state.pd_joint_vel.h5")
dataloader = DataLoader(dataset, batch_size=256, num_workers=0, pin_memory=True, drop_last=True, shuffle=True)
obs, action = dataset[0]
Observation: torch.Size([42])
Action: torch.Size([8])
I get 42 and 8. For the env.reset() however, I get:
obs_mode = "state"
control_mode = "base_pd_joint_vel_arm_pd_joint_vel"
env = gym.make("OpenCabinetDrawer-v1", obs_mode=obs_mode, control_mode=control_mode)
obs = env.reset()
obs.shape (75, )
When I look into the document,
if args.env_id in MS1_ENV_IDS:
if args.control_mode is not None and not args.control_mode.startswith("base"):
args.control_mode = "base_pd_joint_vel_arm_" + args.control_mode
this suggests putting the pd_joint_vel is sufficient to make the control_mode to "base_pd_joint_vel_arm_pd_joint_vel". Is there anything wrong in the order of the codes or how I should format them?
from maniskill.
https://drive.google.com/drive/folders/1oNLgMVkByoHMv6mio-KxTkd7DrE6amKZ
This google drive folder corresponds to LiftCube-v0
, not OpenCabinetDrawer-v1
.
from maniskill.
You can see the list of files here: https://drive.google.com/drive/u/1/folders/1pd9Njg2sOR1VSSmp-c1mT7zCgJEnF8r7
from maniskill.
Related Issues (20)
- [Question/Bug] Size Mismatch for actor_mean Runtime Error HOT 2
- How to take pictures from different cameras? HOT 5
- Record trajectories ray-tracing HOT 7
- [Question] How to load in truly randomized YCB objects each time HOT 3
- Error with motion planning: ManiSkill must have unique names HOT 1
- How to get camera extrinsic and intrinsic parameter? HOT 2
- How to get all environment IDs of ReplicaCAD and AI2THOR? HOT 12
- [Question] Color randomization as in ManiSkillv2 HOT 8
- Cuda kernel init errors caused by massive number of objects being spawned at 0 during reconfiguration
- Regarding the visualization error issue and the parallelization problem in Maniskill3. HOT 2
- Articulation object - bug in get_net_contact_forces with multiple link names HOT 1
- set_qf doesn't appear to control the robot
- [Feature] Add a DAPG baseline HOT 4
- Segmentation fault (core dumped) HOT 4
- Can I scale a loaded actor without changing its shape? HOT 3
- [Bug] joint delta pos to joint pos control conversion has 0 success HOT 3
- Robot control in ArchitecTHOR_SceneManipulation-v1 HOT 4
- Old ManiSkill2 Vulkan Installation Page Broken HOT 1
- [Docs] Modify PPO, SAC, RLPD, and RFCL baselines to report the same metrics where possible for easy comparison
- [Feature] Add in vectorized TDMPC-2 Baseline
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from maniskill.