zemu121

training time

4

When training your model, you use CPU, GPU or multi GPUs? If GPU, what's the version and how long did you take about training your model?

viteribi寻找最佳路径时，为什么使用加法计算： 摘自代码： # 类似隐马模型中的前向概率 alphas = [[[e, None] for e in emissions[0]]] for i in range(len(x) - 1): alphas.append([max([alphas[i][j][0] + transitions[j][k] + emissions[i + 1][k], j] for j in range(4))...

bge-large-zh预训练

5

用自己的数据对bge-large-zh进行预训练，如何在训练过程中对测试集进行loss计算

数据集中只有分类标签，如何计算提取关键词的准确度