search

karpathy / neuraltalk2

Efficient Image Captioning code in Torch, runs on GPU
5.5k stars 1.26k forks source link

When to update the feature embedding layer ? #117

Open Beanocean opened 8 years ago

Beanocean commented 8 years ago

Hi, @karpathy Maybe it is not appropriate to add the feature embedding layer in misc/net_utils.lua( Line 38 ). If the CNN part is not fine-tuned while training, the parameters of feature embedding layer are not updated. What's more the parameters here are initialized randomly.

One more question, How does neuraltalk2 initialize parameters of RNN part ? if I want to initialize them using uniform distribution, can I use code like this: params:uniform(-0.08, 0.08) ?

wakemeupJ commented 7 years ago

I am also confused about this. Maybe we should firstly train a model, then we initialize the parameters of feature embedding layer using the pre-trained model and only train the language model.