servicenow / am3 Goto Github PK

Hi there,

• I have run the code and found that it would take about 12 hours to train 12000 steps on NVIDIA GTX 1080 Ti. Is there anything wrong? Or it would take a long time to train this model indeed.

• And at 12000 steps, the accuracy is no longer improved, the highest is 57.4%. I didn't modify any parameters.Is there anything wrong?

• The program consumes most of the CPU resources but the GPU is rarely used. Also, it only occupies the 159M GPU memory. Is that correct?

Hi, Xing
it's an interesting work, but I wanna , dose the text information only influence the construction of prototype? If so, when there comes visually similar query samples, e.g. a komondor and a mop, their feature embeddings must be very similar(through CNN extracted). In this condition, although the prototype of komondor and mop are dissimilar after considering the text info, but the embeddings of querys with response to above two classes are very similar, hence the final predictions of the two query samples are still very similar and confused.

I don't know whether the text information can affect the learning of CNN.

How is your dataset processed?

Can you tell me how you get the semantic embedding vector for the class label?