alibaba/ai-matrix

DeepInterestNetwork: inconsistant between code and result

imcarsonliao opened this issue · 1 comments

Regarding to deepinterestnetwork, training use batch size of 256,512,1024 and inference use batch size of 1, 32, 64. However, the benchmark result shows 256, 512, 1024 for inference. Is there anything wrong? please help confirm it.

@imcarsonliao there is a typo there. We have fixed that. Please check our new results from version 1.0.2.