lm-scorer
Tyche
lm-scorer | Tyche | |
---|---|---|
4 | 2 | |
294 | 7 | |
- | - | |
0.0 | 2.6 | |
about 2 years ago | about 1 year ago | |
Python | Python | |
MIT License | MIT License |
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.
lm-scorer
- How to obbtain probability for entire sequence (Huggingface transformers)
-
MLM vs CLM for actual language modeling
I've tried this once and found the CLM score to be a better indicator than BERT log prob for my use-case. For CLM, I had used lm-scorer.
- "simonepri/lm-scorer: Language Model based sentences scoring library" ("This package provides a simple programming interface to score sentences using different ML language models.")
-
Whole sentence rather than word frequency nltk?
As in, how generally would a sentence make sense in the totality of English? You could look into language models that give probability of a sentence. You can try a library called lm-scorer.
Tyche
-
[D] With more compute could it be easy to quickly un Mask all the people on Reddit by using text correlations to non masked publicly available text data?
If you’re interested though, I made the following library for my Master’s thesis, which can be used for this: https://github.com/TycheLibrary/Tyche
-
The Green Swan: On the Usefulness of Logic in AI
If you’re interested, the paper I wrote was about a Python library I wrote for belief modelling, called Tyche. I’ve gotten some really promising results with it, although it is still far from the dream of general intelligence.
What are some alternatives?
transformers - 🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
pygraft - Configurable Generation of Synthetic Schemas and Knowledge Graphs at Your Fingertips
penney - Penney's Game
ordered - Entropy-controlled contexts in Python
Sentence-Adder-Anki-Addon - Add sentences to Anki editor window in one click
foodon - The core repository for the FOODON food ontology project. This holds the key classes of the ontology; larger files and the results of text-mining projects will be stored in other repos.
ModuleFormer - ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward experts. We released a collection of ModuleFormer-based Language Models (MoLM) ranging in scale from 4 billion to 8 billion parameters.