MattWallingford/TAPS
Pytorch Implementation of Task Adaptive Parameter Sharing for Multi-Task Learning (CVPR 2022)
When you have a base image recognition model and want to adapt it to several new, related visual tasks—like identifying objects in sketches versus photos—this project helps you do so efficiently. It takes your existing computer vision model and image datasets for new tasks, then fine-tunes a minimal, task-specific part of the model. The output is an adapted model for each new task, ready for specialized image classification, while saving computing resources compared to training separate models from scratch. This is ideal for machine learning engineers or researchers working with image-based AI.
No commits in the last 6 months.
Use this if you need to adapt a computer vision model to multiple distinct image recognition tasks (e.g., classifying sketches, cartoons, or real photos) without retraining the entire model for each task.
Not ideal if your tasks are unrelated or you are building a computer vision model from scratch without a pretrained base model.
Stars
27
Forks
1
Language
Python
License
—
Category
Last pushed
Jul 11, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/MattWallingford/TAPS"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
AdaptiveMotorControlLab/CEBRA
Learnable latent embeddings for joint behavioral and neural analysis - Official implementation of CEBRA
theolepage/sslsv
Toolkit for training and evaluating Self-Supervised Learning (SSL) frameworks for Speaker...
PaddlePaddle/PASSL
PASSL包含 SimCLR，MoCo v1/v2，BYOL，CLIP，PixPro，simsiam, SwAV, BEiT，MAE 等图像自监督算法以及 Vision...
YGZWQZD/LAMDA-SSL
30 Semi-Supervised Learning Algorithms
ModSSC/ModSSC
ModSSC: A Modular Framework for Semi Supervised Classification