A Unity ID allows you to buy and/or subscribe to Unity products and services, shop in the Asset Store and participate in the Unity community.
Separate names with a comma.
https://github.com/Unity-Technologies/ml-agents/releases All of the info about Release 21 is right here. ML-Agents is still in development.
Yes, ordering doesn't matter unless the network contains convolutions. My apologies.
@LastSpark did you end up figuring out how to train on IL2CPP? I know close to nothing about gRPC and have been told that is the primary blocker...
Lots of tickets closed on the github just now - good sign.
Just so everyone who visits this thread is aware, NVIDIA IS making a USD connector which is currently in Beta afaik....
My fix was stack trace logging, link goes to correct answer I found on the forums for my case http://answers.unity.com/answers/1293556/view.html
Thanks for the update
Make sure all control/observational logic is inside FixedUpdateand not Update. There are lots of posts on this forum about people running into...
Make sure the collect observation call isn't happening when your agent game object is set as inactive. If end episode is called before the next...
The problem should be easy enough for a single NN to handle, it's usually about the data, not the NN. There are multiple examples in the ML...
To give credit where credit is due - the old ML-Agents (and Barracuda) team created a really incredible product. This is speaking as a developer...
When it comes to deep learning, hyper parameters can make or break your training sessions. Especially in Reinforcement learning - there are so...
Truly hoping it's because of an IL2cpp integration/major refactor
Make sure that you can drive each individual joint through the heuristic function; you want to emulate the same types of inputs that the NN model...
You can always look at the source code - the ML-Agents team likes to add additional documentation in their code base and you'll often find the...
Interested in this as well
It's common practice in a lot of the ml-agents example projects to utilize a small penalty (-1/maxSteps) to encourage the AI to do anything to...
There's information about the python <--> unity communication pipeline on the github/in the docs. I generally recommend everyone who is using...
I haven't done this myself, though, it sounds super fun. If you go to the post in the sticky section "post your ml-agents project" - a user has...