lin-tan/eagle
For our ICSE22 paper "EAGLE: Creating Equivalent Graphs to Test Deep Learning Libraries" by Jiannan Wang, Thibaud Lutellier, Shangshu Qian, Hung Viet Pham, and Lin Tan.
EAGLE helps deep learning library developers find inconsistencies and bugs in how their libraries process numerical computations. It takes a deep learning library (like TensorFlow or PyTorch) and a version number as input. It then applies a series of transformation rules to numerical operations within the library and reports if these transformations lead to unexpected changes in output, indicating a potential bug. This tool is designed for quality assurance engineers and developers maintaining deep learning frameworks.
No commits in the last 6 months.
Use this if you are a deep learning library developer or QA engineer looking for an automated way to test the numerical stability and correctness of your framework's operations across different versions and configurations.
Not ideal if you are an end-user building or deploying deep learning models and are not involved in the development or maintenance of the underlying deep learning libraries themselves.
Stars
13
Forks
2
Language
Jupyter Notebook
License
—
Category
Last pushed
Aug 16, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/lin-tan/eagle"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
pyg-team/pytorch_geometric
Graph Neural Network Library for PyTorch
a-r-j/graphein
Protein Graph Library
raamana/graynet
Subject-wise networks from structural MRI, both vertex- and voxel-wise features (thickness, GM...
pykale/pykale
Knowledge-Aware machine LEarning (KALE): accessible machine learning from multiple sources for...
dmlc/dgl
Python package built to ease deep learning on graph, on top of existing DL frameworks.