Largest multi-label image database; ResNet-101 model; 80.73% top-1 acc on ImageNet
Role in this project:
ML Engineer Contributions:133 commits, 2 PRs, 29 pushes in 1 year 3 months
Contributions summary:Baoyuan contributed significantly to the data processing and model deployment aspects of the project. They developed scripts for downloading images, converting images into tfrecords, and performed image classification using a pre-trained ResNet model. The commits include modifications to the image downloading pipeline, the creation of tfrecord files for training, and the setup of an image classification script. The user's work also touches on the use of threading for performance and the management of image file lists.
databaseimagenetdeep-learningcomputer-vision
A generic optimization method for any integer programming problem
Contributions:99 commits, 93 pushes, 1 branch in 2 years 7 months
integer-programmingadmmnon-convex-optimizationmachine-learningcomputer-vision