hli1221 / imagefusion_densefuse

DenseFuse (IEEE TIP 2019, Highly Cited Paper) - Python 3.6, TensorFlow 1.8.0
244 stars 62 forks source link

How to feed train data #2

Closed strawberry1996 closed 5 years ago

strawberry1996 commented 5 years ago

Hello, I want to use your code to fusion 10 images into one, but I don't know how to input 10 images , I read your code and I find that in your code , you just input one image . and where is the generate.py used ?

hli1221 commented 5 years ago

You can find 'generate' in 'main.py'. For testing phase, 'IS_TRAINING = False'. And in this 'generate' function, two images are fused.

strawberry1996 commented 5 years ago

Thanks for your answer, I guess that the process of image fusion and training is separate, this right? in other words, when we training , we don't fuse these image, after train, we get the model, then we can use the model to fusion our image.

hli1221 commented 5 years ago

Yes, you are right.

6xfh commented 5 years ago

你好!我是新手,想深入学习您的代码,请问有训练的步骤么?谢谢

hli1221 commented 5 years ago

你好。训练和测试的相关代码入口都在main.py中可以找到。train_recons.py中是训练的主要代码