Do you have the vgg16_no_fc.npy?

DeepRNN / image_captioning

Tensorflow implementation of "Show, Attend and Tell: Neural Image Caption Generation with Visual Attention"

MIT License

785 stars 354 forks source link

Open Kinghup opened 5 years ago

Kinghup commented 5 years ago

Anyone have the file:vgg16_no_fc.npy. The address is invaild. Thank you !

mymuli commented 5 years ago

看第二页，或者百度搜DeepRNN/image_captioning中VGG16和 ResNet50预训练模型

Kinghup commented 5 years ago

已经跑出来了，谢谢~

mymuli commented 5 years ago

你使用python2还是python3 跑的呀？有没有出现json编解码的错误呢？

Kinghup commented 5 years ago

我用两个都跑过，这个里面没遇到，但我遇到过json编解码的问题，自己定义一个encoder就好了

samchild commented 5 years ago

@Kinghup 你有没有训练好的npy呀？可不可以发我一份呀，我的机子训练太慢了，3天才跑完一个batch。我邮箱275569068@qq.com 非常感谢啊最好是python3的哈~~

463728946 commented 5 years ago

@Rmuli 百度搜索的连接也失效了，你说的第二页是哪里？

463728946 commented 5 years ago

@Kinghup 你是怎么跑出来的？

caoquanjie commented 5 years ago

@Kinghup 我跑出来的bleu1的结果没上70 请问你跑出来的和作者的一样能达到70以上吗如果达到了方便回复我一下吗十分感谢