tf2报错解决 assertion failed: [predictions must be ＞= 0] [Condition x ＞= y did not hold element-wise:]

环境： tf2

完整报错如下：

tensorflow.python.framework.errors_impl.InvalidArgumentError:  assertion failed: [predictions must be >= 0] [Condition x >= y did not hold element-wise:] [x (sub_2:0) = ] [[-nan][-nan][-nan]...] [y (Cast_2/x:0) = ] [0]
         [[{{node assert_greater_equal/Assert/AssertGuard/else/_1/assert_greater_equal/Assert/AssertGuard/Assert}}]] [Op:__inference_train_function_1780]

解决方法：

1. 检查模型最后一层，是否忘记添加激活函数softmax

layers.Dense(number_of_tags, activation='softmax', name='decoder_dense')

1. 检查是否调用了sqrt这类的函数，并要确保目标不为零

tf.sqrt(xxx+tf.keras.backend.epsilon)

1. 检查learning rate是否过大，导致nan