search

lambdaji / tf_repos

TensorFlow Script
681 stars 318 forks source link

RuntimeError: There was no new checkpoint after the training. Eval status: no new checkpoint #27

Open xxllp opened 5 years ago

xxllp commented 5 years ago

模型训练都会报这个错，谁知道咋解决啊

lambdaji commented 4 years ago

应该是模型或者batch比较大，还没计算完eval已经开始了，调整下 EvalSpec的参数

Melaniemod commented 4 years ago

而且要在命令行中运行而不是在pycharm中运行