xiaocode-er / two-stream-cnn Goto Github PK

good work!
would you mind sharing your pre-trained model. I would like to test the result, my model not work well as you metioned

我之前利用keras自带的Imagegenerator函数，想用Imagegenerator.flow_from_directory（）来处理两组图像，然后使用model.fit.generator（）函数对模型进行拟合，但是由于keras自的.fit.generator（）函数固定只能输入一个generator（尝试改了源代码但是一直报错），所以未能成功。很幸运找到了这个项目，想在您的基础上修改定义一个（训练和测试分别都）送入两组图片的函数，看您用train_on_batch（）可以实现多输入，我想请问一下您应该怎么根据我的实际情况来定义自己的data_generator呢？（万分感激！）

I would like to retrain your model on my dataset and check the results. May i know how can i train on my own dataset. what would be the format of the dataset

Hi can you send me the cs_trn.hdf5 and cs_test.hdf5 files
my email [email protected]

Hi, thx for your outstanding work!
I notice that there is no data normalization in your code
and there is no bn layers, too.
Could you explain it?
:)

@XiaoCode-er
Hello, thanks for your outstanding work!

It will be better to release you /*.hdf5 file in you code.

Thank you very much!

Hi ,I notice that you introduce attention mechanism to your network, could you send the attention to me ? I want to introduce attention to my network too, I want to learn from your code. Thank you for your help. @XiaoCode-er

Thanks for your amazing work!
Here comes one question: in NTU, some videos contain only one person, but some may contain two or more persons. So I am confused about how you solve this.
I read your code, which shows you only solve two-person situation. Am I right?

Can you please share with us the prediction script?