Leo

Thank you for answering. I would like to ask, when training a recognition model, does the model learn to recognize space when feeding input images containing only one word. Would...

within one week, I need to train it with larger dataset which takes longer time

in the triplet framework, a batch of training data consists of 3 parts: (anchor, pos, neg) positive_distance means the l2 distances between anchor and pos. anchor and pos are embeddings...

if all the layers is learnable, try a smaller learning rate. For the first time, I usually froze all the layers except the last layer, train the last layer with...

@antipink try to use the model to predict somthing, if the prediction looks wierd (for example, the network predicts the same vector no matter what input is), then the learning...

this should be a fine-tuning process, it assumes that a pre-trained face network (vgg-face) was loaded before training. I dont think initializing an empty vgg or image-net vgg would work...