google-research
ml-agents
google-research | ml-agents | |
---|---|---|
98 | 60 | |
32,915 | 16,394 | |
1.1% | 1.2% | |
9.6 | 8.0 | |
3 days ago | 8 days ago | |
Jupyter Notebook | C# | |
Apache License 2.0 | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
google-research
-
Show HN: Next-token prediction in JavaScript – build fast LLMs from scratch
People on here will be happy to say that I do a similar thing, however my sequence length is dynamic because I also use a 2nd data structure - I'll use pretentious academic speak: I use a simple bigram LM (2-gram) for single next-word likeliness and separately a trie that models all words and phrases (so, n-gram). Not sure how many total nodes because sentence lengths vary in training data, but there are about 200,000 entry points (keys) so probably about 2-10 million total nodes in the default setup.
"Constructing 7-gram LM": They likely started with bigrams (what I use) which only tells you the next word based on 1 word given, and thought to increase accuracy by modeling out more words in a sequence, and eventually let the user (developer) pass in any amount they want to model (https://github.com/google-research/google-research/blob/5c87...). I thought of this too at first, but I actually got more accuracy (and speed) out of just keeping them as bigrams and making a totally separate structure that models out an n-gram of all phrases (e.g. could be a 24-token long sequence or 100+ tokens etc. I model it all) and if that phrase is found, then I just get the bigram assumption of the last token of the phrase. This works better when the training data is more diverse (for a very generic model), but theirs would probably outperform mine on accuracy when the training data has a lot of nearly identical sentences that only change wildly toward the end - I don't find this pattern in typical data though, maybe for certain coding and other tasks there are those patterns though. But because it's not dynamic and they make you provide that number, even a low number (any phrase longer than 2 words) - theirs will always have to do more lookup work than with simple bigrams and they're also limited by that fixed number as far as accuracy. I wonder how scalable that is - if I need to train on occasional ~100-word long sentences but also (and mostly) just ~3-word long sentences, I guess I set this to 100 and have a mostly "undefined" trie.
I also thought of the name "LMJS", theirs is "jslm" :) but I went with simply "next-token-prediction" because that's what it ultimately does as a library. I don't know what theirs is really designed for other than proving a concept. Most of their code files are actually comments and hypothetical scenarios.
I recently added a browser example showing simple autocomplete using my library: https://github.com/bennyschmidt/next-token-prediction/tree/m... (video)
And next I'm implementing 8-dimensional embeddings that are converted to normalized vectors between 0-1 to see if doing math on them does anything useful beyond similarity, right now they look like this:
[nextFrequency, prevalence, specificity, length, firstLetter, lastLetter, firstVowel, lastVowel]
- Google Research website is down
-
Jpegli: A New JPEG Coding Library
The change was literally just made: https://github.com/google-research/google-research/commit/4a...
It appears this was in response to Hacker News comments.
- Multi-bitrate JPEG compression perceptual evaluation dataset 2023
-
Vector Databases: A Technical Primer [pdf]
There are options such as Google's ScaNN that may let you go farther before needing to consider specialized databases.
https://github.com/google-research/google-research/blob/mast...
-
Labs.Google
I feel it was unnecesary to create this because https://research.google/ already exists? It just seems like they want to take another URL with a "pure" domain name instead of psubdirectories, etc parts.
-
Smerf: Streamable Memory Efficient Radiance Fields
https://github.com/google-research/google-research/blob/mast...
-
Shisa 7B: a new JA/EN bilingual model based on Mistral 7B
You could also try some dedicated translation models like https://huggingface.co/facebook/nllb-moe-54b (or https://github.com/google-research/google-research/tree/master/madlad_400 for something smaller) and see how they do.
-
Translate to and from 400+ languages locally with MADLAD-400
Google released T5X checkpoints for MADLAD-400 a couple of months ago, but nobody could figure out how to run them. Turns out the vocabulary was wrong, but they uploaded the correct one last week.
- Mastering ROUGE Matrix: Your Guide to Large Language Model Evaluation for Summarization with Examples
ml-agents
- How do I change the maximum number of steps for training
- are the install steps update to date?
-
Help with regenerating new worker id
I am a beginner to using ML Agents to simulate an environment for DL i am trying to trial runs by tinkering through different values between the action space and keep encountering this issue when attempting to run a new trial. I've tried mlagents-learn --force and mlagents-learn --run-id=newtest but both prompt the same error message. Using linux, I am aware of a similar bug occuring in older versions (https://github.com/Unity-Technologies/ml-agents/issues/1505) but solutions didn't fix it.
-
Trying to get into AI
The Github page for ML-Agents has a fairly straight forward example.
- Implement API to allow AI/ML to play your game, or is it not needed?
-
Is there a good alternative to Unity ML Agents?
Very few commits in the last year and not many new features (https://github.com/Unity-Technologies/ml-agents/commits/develop)
-
At least I put effort into the AI prompt to generate some code that people can refer to, whereas you do absolutely nothing to contribute to the community.
and PR content: https://github.com/Unity-Technologies/ml-agents/commit/ed212103e451449bf84711a4a8f7bf11dfb1211a
-
I have some questions as an absolute beginner.
Unity can build a stand-alone application or be used as a library. Javascript is deprecated, and Boo along with it although it was never really supported to begin with. Various types of machine learning are supported through the ML-Agent Toolkit and pretty well documented. The toolkit has a Python API but you should be careful about doing anything too unusual in Unity because the documentation tends to have a lot of dead-ends.
-
Could Somebody please help me figure this out ? been struggling with it for a week now
Op, I'd just pull the repo again to a new folder from https://github.com/Unity-Technologies/ml-agents (use SourceTree for simplicity if you don't know git).
-
Unity ML-Agents documentation is wrong, I can't build an executable and run training as the docs state
My github issue on their documentation: https://github.com/Unity-Technologies/ml-agents/issues/5899
What are some alternatives?
qdrant - Qdrant - High-performance, massive-scale Vector Database for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
gym - A toolkit for developing and comparing reinforcement learning algorithms.
fast-soft-sort - Fast Differentiable Sorting and Ranking
AirSim - Open source simulator for autonomous vehicles built on Unreal Engine / Unity, from Microsoft AI & Research
faiss - A library for efficient similarity search and clustering of dense vectors.
carla - Open-source simulator for autonomous driving research.
Milvus - A cloud-native vector database, storage for next generation AI applications
AssetStudio - AssetStudio is a tool for exploring, extracting and exporting assets and assetbundles.
struct2depth - Models and examples built with TensorFlow
unity-avatar-generation - A minimal example of how to use Unity's AvatarBuilder.BuildHumanAvatar API.
bootcamp - Dealing with all unstructured data, such as reverse image search, audio search, molecular search, video analysis, question and answer systems, NLP, etc.
ultimate-volleyball - 3D RL Volleyball environment built on Unity ML-Agents