GitHub - earhian/Humpback-Whale-Identification-1st-: https://www.kaggle.com/c/hu...
source link: https://github.com/earhian/Humpback-Whale-Identification-1st-
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
readme.md
Kaggle Humpback Whale Identification Challenge 1st place code
Recent Updates
[2019.3.1 16:44] uploading mask file
Dependencies
- python==3.6
- torch==0.4.1
- torchvision==0.2.1
Solution
https://www.kaggle.com/c/humpback-whale-identification/discussion/82366
prepare data
competition dataset
trainset -> ./input/train
testset -> ./input/test
mask
cd input
unzip mask.zip
download https://drive.google.com/file/d/1hfOu3_JR0vWJkNlRhKwhqJDaF3ID2vRs/view?usp=sharing into ./input
playground data
download playground data, then put them into input/train
https://www.kaggle.com/c/whale-categorization-playground/data
Train
line 301 in train.py
step 1.
freeze = False
model_name = 'senet154'
min_num_class = 10
checkPoint_start = 0
lr = 3e-4
#until train map5 >= 0.98
step 2.
freeze = True
model_name = 'senet154'
min_num_class = 0
checkPoint_start = best checkPoint of step 1
lr = 3e-4
step 3.
freeze = True
model_name = 'senet154'
min_num_class = 0
checkPoint_start = best checkPoint of step 2
lr = 3e-5
Test
line 99 in test.py
checkPoint_start = best checkPoint of step 3
Recommend
About Joyk
Aggregate valuable and interesting links.
Joyk means Joy of geeK