Skip to content

poeticmage/Notebook-data-pipeline

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 

Repository files navigation

Notebook-data-pipeline

This repository contains jupyter notebooks of data preprocessing of various datasets before being fed to the model. Proper train,test split, class handling and train test loaders are written in code. This is to help those who face difficulty in the preliminary step. A bit of guidance would surely help many and save effort and time

Awa2 (Animals with Attributes2)

This dataset is primarily used in Zero-shot Learning. The split that is used is the proposed split by Xian et al. in Zero Shot Learning -- the good, the bad and the ugly, so that testing images are not used in training the large backbones people often use. Reference: C. H. Lampert, H. Nickisch, and S. Harmeling. "Learning To Detect Unseen Object Classes by Between-Class Attribute Transfer". In CVPR, 2009

CUB-200 (Caltech-UCSD Birds 200)

This dataset is used in Zero-shot Learning. The notebook proposes a pipeline to prepare train and test loader for implementation in model training. The first 50 are used as test(unseen) data and rest 150 as train classes. Reference: Wah, C., Branson, S., Welinder, P., Perona, P., & Belongie, S.:The Caltech-UCSD Birds-200-2011 Dataset, California Institute of Technology, 2011.

About

This repository contains jupyter notebooks of data preprocessing of various datasets before being fed to the model. Proper train,test split, class handling and train test loaders are written in code. This is to help those who face difficulty in the preliminary step. A bit of guidance would surely help many and save effort and time

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors