StonyBrookNLP/irene
[ACL 2021] IrEne: Interpretable Energy Prediction for Transformers
This project helps machine learning researchers and engineers predict the energy consumption of large language models (specifically Transformer models). You input details about a Transformer model, including its architecture, batch size, and sequence length, and it outputs an estimate of the energy usage in Joules for the overall model and its individual components. This is useful for those aiming to optimize the energy efficiency of their NLP applications.
No commits in the last 6 months.
Use this if you need to understand and predict the energy consumption of different Transformer models and their internal operations, helping you choose more energy-efficient architectures or deployment settings.
Not ideal if you are looking to measure the energy consumption of non-Transformer models or hardware beyond the specified devices, or if you need real-time energy monitoring.
Stars
11
Forks
—
Language
Python
License
—
Category
Last pushed
Sep 08, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/StonyBrookNLP/irene"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
rmovva/HypotheSAEs
HypotheSAEs: hypothesizing interpretable relationships in text datasets using sparse...
interpretml/interpret-text
A library that incorporates state-of-the-art explainers for text-based machine learning models...
fdalvi/NeuroX
A Python library that encapsulates various methods for neuron interpretation and analysis in...
jalammar/ecco
Explain, analyze, and visualize NLP language models. Ecco creates interactive visualizations...
alexdyysp/ESIM-pytorch
中国高校计算机大赛--大数据挑战赛