Is data from kaggle different from here?
kxhit opened this issue · 2 comments
Hi! I find another source of data from kaggle https://www.kaggle.com/c/landmark-retrieval-2020/data
What's the relation betweeen kaggle data/csv with data/csv in this repo? I'm confused. Thanks if you could give some explanation.
In my opinion, Kaggle basically allows you to find and publish datasets and also the csv, it can be imported or exported using programs that stores data in tables.
The one listed in this repo is the 100% official/complete version.
In Kaggle, in some cases the data may have been subsampled/resized, depending on the setting. (for example, in the pointer you gave, it shows the GLDv2-clean version of the training set -- The training data for this competition comes from a cleaned version of the Google Landmarks Dataset v2 (GLDv2)
)