How to re-produce dbm multimodal model
sunbaigui opened this issue · comments
I'm very impressed by your work Multimodal Learning with Deep Boltzmann Machines(http://www.cs.toronto.edu/~nitish/multimodal/). But when I wanna try to re-produce this model by following your code https://github.com/nitishsrivastava/deepnet, I find there's hardly a tutorial about how to train the multimodel dbm model referenced in your paper. @nitishsrivastava
Have you solved your problem? I have the same question in training multi-modal DBM
I just use the feature extracted by the joint layer,anyway,if you want to reproduce the bimodal DBM,you can just add another layer after the joint layer and train it just like before
Hi,
I have successfully trained the model by running ./runall_dbn.sh. However I do not know to use it. How do I generate the text tags for an image? Please help.
Thank you,
Sreekar Reddy.
