未分類

BirdCLEF 2021 – Birdcall Identification Participation Record

スポンサーリンク

Kaggle’s first competition

Although I registered for this competition three years ago, this will be my first time actually participating. For now, I decided to participate because it was the shortest deadline for points and medals.

Purpose of this competition

Categorize what kind of bird it is by its call

Current Status

I submitted the highest rate kernel as is, and got 0.61. I heard that the score is from the 35% part of the total, and the score can change depending on the 65% part. I don’t know how many of them there are, but I guess it means 61% correct answers. But there is no room for improvement that I can make.

What to do

1) Improve basic skills (just from Titanic and whatnot)
2) Understand what the submitted kernels and basic samples are doing.
3) Improve your basic skills (rather than before the deadline)

4) Try to do your best somehow with the knowledge you have acquired.

Goals

Bronze medal range (how can that be sweet…)

What I’m doing

The kernel is now 0.64, so I copied and pasted and submitted. I think I tweaked some parameters?
Bought Bayesian Statistics from udemy as the course is sale.
I’ve been looking for something to learn about machine learning, and I found a course on kaggle that goes into a lot of depth, so I took it. I’m not sure if this is a good idea or not, but it’s a good idea. It’s a great time.

コメント

タイトルとURLをコピーしました