walsvid/CoordConv
Pytorch implementation of "An intriguing failing of convolutional neural networks and the CoordConv solution" - https://arxiv.org/abs/1807.03247
This project helps deep learning engineers improve how their convolutional neural networks (CNNs) handle spatial relationships in data. It takes standard PyTorch convolutional layers and augments them to include coordinate information, which can lead to more accurate predictions, especially for tasks sensitive to an object's position. This is for machine learning engineers and researchers building and training CNN models.
163 stars. No commits in the last 6 months.
Use this if your convolutional neural networks struggle with tasks that require precise understanding of an object's location or its relationship to other elements within an image or sequence.
Not ideal if you are working with non-spatial data, or if your current CNNs already achieve satisfactory performance on coordinate-sensitive tasks.
Stars
163
Forks
28
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Jan 08, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/walsvid/CoordConv"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
Jittor/jittor
Jittor is a high-performance deep learning framework based on JIT compiling and meta-operators.
berniwal/swin-transformer-pytorch
Implementation of the Swin Transformer in PyTorch.
zhanghang1989/ResNeSt
ResNeSt: Split-Attention Networks
NVlabs/FasterViT
[ICLR 2024] Official PyTorch implementation of FasterViT: Fast Vision Transformers with...
ViTAE-Transformer/ViTPose
The official repo for [NeurIPS'22] "ViTPose: Simple Vision Transformer Baselines for Human Pose...