floodsung / wechat_jump_end_to_end_train Public

Notifications You must be signed in to change notification settings
Fork 14
Star 45

Playing Wechat Jump Game with End-to-End Convolutional Neural Networks (Train code)

45 stars 14 forks Branches Tags Activity

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 5 Commits
LICENSE		LICENSE
README.md		README.md
cnn.pkl		cnn.pkl
dataset.py		dataset.py
dataset84.py		dataset84.py
model.pkl		model.pkl
run_mask.py		run_mask.py
run_vgg.py		run_vgg.py
train.py		train.py
train_mask.py		train_mask.py
train_vgg.py		train_vgg.py
unet.pkl		unet.pkl
unet.py		unet.py
vgg.py		vgg.py

Repository files navigation

Wechat_Jump_End_to_End_Train

Prerequisite

pytorch

How to use it

Download the dataset from Baidu Yun

Place the dataset in this folder, then

python train.py

I train this model on Ubuntu 16.04 with GTX 970

Train vgg version and attention version

Almost the same with the basic version. Just try it.

Reference

https://github.com/wangshub/wechat_jump_game

About

Playing Wechat Jump Game with End-to-End Convolutional Neural Networks (Train code)

Report repository

Releases

No releases published

Packages

No packages published

Languages

Python 100.0%