Skip to content

ModerateHatespeech/model

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

26 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Machine Learning Models

In the interest of research and contribution into the performance of our language models, we release previous versions of our models for public use.

License

All models are licensed under the GPL-v3 license, and released on an as-is basis. We make no guarantee of the accuracy or performance of these models, and you implement them under the expectation that these models can potentially contain biases as a result of the training process.

Download

wget https://moderatehatespeech.com/research/models/pyt_v2.pkl -O production.pkl

Usage

The following weights are provided as FastAI (Pytorch) pickled weights. You should install requirements via: pip3 install -r requirements.txt To load the model, import the necessary classes from the mhs_imports.py file and load the model:

from mhs_imports import *
model = MHSPredictor("production.pkl")

To run a inference:

model.predict("Testing, 123")

Training

Information on our training process can be found on our website

About

Machine learning models for Toxicity detection

Topics

Resources

License

Stars

Watchers

Forks

Contributors

Languages