At inference time, we need to handle the case when there is a turtle image fro, unseen class (which is not in the 100 classes on which we train). How do we handle this case?
you can images from extra.csv and treat them as "new_turtle". it will become a classification challenge but you need to double check the added images before training.
I tried adding images from `extra_images.csv` and assigned images with new ids with "turtle_id". It helps increase my CV score but I'm getting a huge drop in the leaderboard score.
how many images from `extra_images.csv` did you add ? make sure to keep the same distribution of the train images. try adding around 20 images from the extra to train and see the results. Good Luck!
'/%3e%3c/defs%3e%3cpath fill='%23fff' d='M-120-80h240V80h-240z'/%3e%3cpath d='M-120-80h240v48h-240z'/%3e%3cpath fill='%23060' d='M-120 32h240v48h-240z'/%3e%3cg id='b'%3e%3cuse xlink:href='%23a' stroke='%23000'/%3e%3cuse xlink:href='%23a' fill='%23fff'/%3e%3c/g%3e%3cuse xlink:href='%23b' transform='scale(-1 1)'/%3e%3cpath fill='%23b00' d='M-120-24v48h101c3 8 13 24 19 24s16-16 19-24h101v-48H19C16-32 6-48 0-48s-16 16-19 24z'/%3e%3cpath id='c' d='M19 24c3-8 5-16 5-24s-2-16-5-24c-3 8-5 16-5 24s2 16 5 24'/%3e%3cuse xlink:href='%23c' transform='scale(-1 1)'/%3e%3cg fill='%23fff'%3e%3cellipse rx='4' ry='6'/%3e%3cpath id='d' d='M1 5.85s4 8 4 21-4 21-4 21z'/%3e%3cuse xlink:href='%23d' transform='scale(-1)'/%3e%3cuse xlink:href='%23d' transform='scale(-1 1)'/%3e%3cuse xlink:href='%23d' transform='scale(1 -1)'/%3e%3c/g%3e%3c/svg%3e)
One Idea would be to use probability thresholding
How is that idea compared to adding an "unknown" class with images not contained in the 100 classes?
you can images from extra.csv and treat them as "new_turtle". it will become a classification challenge but you need to double check the added images before training.
I tried adding images from `extra_images.csv` and assigned images with new ids with "turtle_id". It helps increase my CV score but I'm getting a huge drop in the leaderboard score.
how many images from `extra_images.csv` did you add ? make sure to keep the same distribution of the train images. try adding around 20 images from the extra to train and see the results. Good Luck!
Thanks. I tried sampling 20-30 images at random multiple times but still getting a drop of 2-3% in my leaderboard score.