Skip to content

A solution to the Carvana Image Masking Challenge on Kaggle (based on tensorflow)

License

Notifications You must be signed in to change notification settings

wangheda/Carvana-NoOne

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

kaggle-carvana

Code for the competition Carvana on Kaggle

System Requirement

Not all requirement are needed, but python should be 2.x and tensorflow should be 1.0+. Also, the scripts in this repository may not work properly on windows.

Ubuntu 16.04
Nvidia CUDA 8.0 with driver version 378.13
Python 2.7
Tensorflow 1.2
Sklearn 0.18.1
Scikit-image 0.12.3

data conversion

There are some conventions about where we put the original files. (If the quota is not enough, use a soft link.)

Images: train-images/*.jpg
Labels: train-masks/*.gif
Test: test-images/*.jpg

There are also some conventions about where the tfrecord files are put to. Make sure you have the writing permissions.

Train: train-data/data*.tfrecord
Test: test-data/data*.tfrecord

RUN this script to generate tfrecords files.

bash convert-image-into-tfrecords.sh 

About

A solution to the Carvana Image Masking Challenge on Kaggle (based on tensorflow)

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published