Yuhao Lu
Yuhao Lu
another option: add these 2 lines before import flair `import conllu` `conllu.TokenList = conllu.models.TokenList`
I put the memory forward and backward on CPU, when the memory became large. it was a little bit slower but worked well. model_forward_gpu -> memory_forward_cpu -> loss_cpu -> memory_backward_cpu...
公司:图灵机器人 网址:http://www.turingapi.com/ 地点:北京市海淀区 联系方式:[email protected] 使用场景:向量检索
https://github.com/pallashadow/SuperGlue-pytorch/blob/master/models/superglueLoss.py @Shuhul24 @SaharAlmahfouzNasser