level 1
little光1
楼主
小白提问，do_train=true没问题，请问下bert在do_eval=true时出现以下的错误是什么原因，希望能得到建议或者解决方案
ERROR:tensorflow:Error recorded from evaluation_loop: 2 root error(s) found.
(0) Invalid argument: assertion failed: [predictions must be in [0, 1]] [Condition x <= y did not hold element-wise:x (ArgMax:0) = ] [0 0 2...] [y (auc/Cast_1:0) = ] [1]
[[node auc/assert_less_equal/Assert/Assert (defined at /zhangzhaoguang/run_classifier.py:728) ]]
[[accuracy/broadcast_weights/assert_broadcastable/AssertGuard/Assert/data_7/_1326]]
(1) Invalid argument: assertion failed: [predictions must be in [0, 1]] [Condition x <= y did not hold element-wise:x (ArgMax:0) = ] [0 0 2...] [y (auc/Cast_1:0) = ] [1]
[[node auc/assert_less_equal/Assert/Assert (defined at /zhangzhaoguang/run_classifier.py:728) ]]
2020年08月14日 08点08分
1
ERROR:tensorflow:Error recorded from evaluation_loop: 2 root error(s) found.
(0) Invalid argument: assertion failed: [predictions must be in [0, 1]] [Condition x <= y did not hold element-wise:x (ArgMax:0) = ] [0 0 2...] [y (auc/Cast_1:0) = ] [1]
[[node auc/assert_less_equal/Assert/Assert (defined at /zhangzhaoguang/run_classifier.py:728) ]]
[[accuracy/broadcast_weights/assert_broadcastable/AssertGuard/Assert/data_7/_1326]]
(1) Invalid argument: assertion failed: [predictions must be in [0, 1]] [Condition x <= y did not hold element-wise:x (ArgMax:0) = ] [0 0 2...] [y (auc/Cast_1:0) = ] [1]
[[node auc/assert_less_equal/Assert/Assert (defined at /zhangzhaoguang/run_classifier.py:728) ]]
