3rd place solution
#16
by
fconti
- opened
Hi!
I have created a repository on GitHub that includes the following:
- A technical report that discusses the approach in detail.
- An evaluation notebook that allows you to reproduce the CSV file using the trained model (available for download from Google Drive).
- A training notebook that enables you to replicate the training process using the triplet loss approach.
I hope you'll enjoy my approach! If you spots any errors or issues, or if you have any suggestions for improvement, please don't hesitate to let me know!
Thanks for sharing.