asurada

Hi, I'm also training the network from scratch using the pre-trained vgg16, but I can't get the same result. Did you used the pre-trained vgg16? @Mahsa13473

Download [vgg_16.ckpt](http://download.tensorflow.org/models/vgg_16_2016_08_28.tar.gz) and save to `./models/CNN/pretrained_model` first. @JohnG0024

The VGG was used as an encoder to extract the features of the image. The pre-trained VGG was training on [ImageNet](http://www.image-net.org/) dataset(more than 14 million images and more than 20,000...

You can find more details in this [paper](https://arxiv.org/abs/1409.1556) @JohnG0024

Thanks a lot! This work for me.