zswang666 / softzoo Goto Github PK
View Code? Open in Web Editor NEWLicense: MIT License
License: MIT License
Hi! I am trying to use multiple environments in my program, like the following code:
args = [setting1, setting2, setting3]
for i in range(3):
env = make_env(args[i])
# environmental interactions and training process
When I switch to the next environmental setting, I use ti.reset() to successfully release GPU memory (this works, thx), but I found that the cpu memory of the program still increase (this might because the program needs to execute env.initialize() again). Is there any way to free up cpu memory occupied by the previous environment? Otherwise, the computer's memory will overflow.
Thx!
Hi! When I want to use multiple environments in my program, like the following code:
TRAINENVS=["ground", "wetland", "ice", "shallow_water"]
for i in range(len(TRAINENVS)):
ti.init(arch=ti.gpu,device_memory_fraction=0.8)
individual_args={"env_config_file": f"{TRAINENVS[i]}.yaml",
"annotated_pcd_path": f"./softzoo/assets/meshes/pcd/panda.pcd"}
for k, v in individual_args.items():
setattr(args, k, v)
env = make_env(args)
# Doing some training under this env
When I finish a training round and start training in the next environment, I found that the memory of the program increases. Is there any way to free up memory occupied by a previous environment? Otherwise the computer's memory will overflow.
Thx!
Hi! In SoftZoo, there are FinalStepCoMLoss, TrajectoryFollowingLoss, PerStepCoVLoss, AnimatedEMDLoss, VelocityFollowingLoss, WaypointFollowingLoss, RotationLoss, ThrowingObjectLoss and ObjectVelocityLoss to be selected for a specific task. I don't know how to choose a correct loss. Can you provide some descriptions of these loss functions?
Thx!
Why are you so good?
pip install -r requirments.txt -> pip install -r requirements.txt
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.