Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Great Barrier Reef Model Pipeline: 15th place
Search
Maxwell
February 16, 2022
Science
1
140
Great Barrier Reef Model Pipeline: 15th place
https://www.kaggle.com/c/tensorflow-great-barrier-reef
All I want to use was YOLO-X!
Maxwell
February 16, 2022
Tweet
Share
More Decks by Maxwell
See All by Maxwell
Causal Impact -paper summary-
hoxomaxwell
2
520
Lecture materials at the University of Tokyo School of Medicine
hoxomaxwell
1
89
Kaggle Hungry Geese
hoxomaxwell
1
74
HuBMAP 17th place model pipeline
hoxomaxwell
1
65
LT: Shallow Dive into Bayes Factor
hoxomaxwell
6
1.2k
Kaggle APTOS 2019 @ U-Tokyo Med
hoxomaxwell
1
390
Cornell Birdcall 36th place solution
hoxomaxwell
2
210
Kaggle Bengali.AI 6 th place solution
hoxomaxwell
4
7.7k
Google Colaboratory Shortcuts
hoxomaxwell
2
970
Other Decks in Science
See All in Science
Snowflakeによる統合バイオインフォマティクス
ktatsuya
0
200
プロダクト開発を通して学んだナレッジマネジメントの哲学
sonod
0
110
「国と音楽」 ~spotifyrを用いて~ #muana
bob3bob3
2
390
汎用原子シミュレータMatlantis のご紹介
matlantis
0
270
HAS Dark Site Orientation
astronomyhouston
0
5.1k
遺伝子発現プロファイルに基づく新しい薬物間相互作用予測法
tagtag
0
130
効果検証入門に物申してみた_JapanR_2023
s1ok69oo
6
4.8k
様々な侵入者タイプに対応した適切な警備計画の策定 / Patrol route design considering various types of intrudes
konakalab
0
120
Running llama.cpp on the CPU
ianozsvald
0
310
Leveraging conformal prediction for calibrated probabilistic time series forecasts to accelerate the renewable energy transition
ingevandenende
2
310
Machine Learning for Materials (Lecture 4)
aronwalsh
0
730
深層学習を利用して 大豆の外部欠陥を判別した研究事例の紹介
kentaitakura
0
150
Featured
See All Featured
Product Roadmaps are Hard
iamctodd
PRO
48
10k
Code Reviewing Like a Champion
maltzj
517
39k
Fontdeck: Realign not Redesign
paulrobertlloyd
79
5.1k
It's Worth the Effort
3n
181
27k
How to Ace a Technical Interview
jacobian
274
23k
CSS Pre-Processors: Stylus, Less & Sass
bermonpainter
353
29k
What's in a price? How to price your products and services
michaelherold
239
11k
KATA
mclloyd
20
13k
Agile that works and the tools we love
rasmusluckow
325
20k
Gamification - CAS2011
davidbonilla
78
4.9k
Statistics for Hackers
jakevdp
792
220k
Exploring the Power of Turbo Streams & Action Cable | RailsConf2023
kevinliebholz
12
3.8k
Transcript
Copyright 2022 Maxwell_110 Validation strategy - Sequence-based 4 fold CV
- The number of CoTS is close in each fold - Training data is frames with CoTs - Validation data includes frames w/o CoTs Resize up to 2.75 times using progressive learning 1280 720 Augmentation Increasing probability of applying augmentation as progressive learning progresses. - Default YOLO-X augmentations - random resize: (-5, 5) - mosaic / MixUp / hsv / flip: p = 0.6 -> 0.8 - degrees: Not used - translate: 0.1 - mosaic / MixUp scale: (0.5, 1.5) - RandomGamma - RGBShift - Sharpen - GaussNoise Batch Size: 4 GeForce RTX 3080 (x 2) Solution description in Kaggle discussion https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307691 Learning strategy - Progressive learning - Optimizer: default SGD (decay: 5e-4, momentum: 0.9) - LR: .000625 - Scheduler: yoloxwarmcos - min_lr_ratio: 0.1 - EMA: on - warmup_epochs: 5 - max_epoch: 30 TTA Seq-NMS https://arxiv.org/abs/1602.08465 https://github.com/tmoopenn/seq-nms n_frames: 2 confidence threshold: 0.07 linkage threshold: 0.1 nms th: 0.4 Weighted Box Fusion skip box threshold: 0.05 wbf IoU threshold: 0.45 Final confidence threshold: .08 Public LB : 0.607 Private LB : 0.714