dbaranchuk/memory-efficient-maml
Memory efficient MAML using gradient checkpointing
This project helps machine learning engineers or researchers train complex meta-learning models more efficiently. It takes a PyTorch model and MAML training configuration as input, enabling the execution of many more meta-learning steps without exceeding GPU memory limits. The output is a meta-learned model that can adapt quickly to new, unseen tasks.
No commits in the last 6 months.
Use this if you are a machine learning practitioner working with Model-Agnostic Meta-Learning (MAML) and are running into GPU memory constraints when attempting to use many MAML steps.
Not ideal if you are not using PyTorch for your deep learning models or if your primary bottleneck is not GPU memory during MAML training.
Stars
86
Forks
7
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Dec 30, 2019
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/dbaranchuk/memory-efficient-maml"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
jakesnell/prototypical-networks
Code for the NeurIPS 2017 Paper "Prototypical Networks for Few-shot Learning"
harveyslash/Facial-Similarity-with-Siamese-Networks-in-Pytorch
Implementing Siamese networks with a contrastive loss for similarity learning
oscarknagg/few-shot
Repository for few-shot learning machine learning projects
google-research/meta-dataset
A dataset of datasets for learning to learn from few examples
Sha-Lab/FEAT
The code repository for "Few-Shot Learning via Embedding Adaptation with Set-to-Set Functions"