Results
2
issues of
作小寒
Hello author, I used the image caption model you provided, the accuracy is good, but the inference is slow. I saw that there is a transformer-based nlpconnect/vit-gpt2-image-captioning model that computes...
修改代码，归一化数据