FuLin

+1 Tensorboard 2.14 on AWS SageMaker notebook instance. doesn't work

Same problem on version 1.8

Solved the problem. Key is to notice any permission error when deploying the training and serving stack. bash docker_deploy.sh all --training bash docker_deploy.sh all --serving

增加词表长到20k，TextRNN的准确率我这能涨接近2个点

现已加入对Bert的支持

应该是参考了Bert原文中Feature-based Approach with BERT的部分

应该是参考了Bert原文中Feature-based Approach with BERT的部分

+1 同样的问题

我这边后来发现的原因是Fate serving没有正确启动