問題：在使用keras-bert 導入預訓練的模型時， 報錯Layer model_1 expects 3 inputs, but it received 2 input tensors
導入代碼：
bert_model = load_trained_model_from_checkpoint(config_path, checkpoint_path, training=True,output_layer_num=7,trainable=True,seq_len=Config.max_seq_len) # 加載預訓練模型
具體報錯：
Layer model_1 expects 3 inputs, but it received 2 input tensors. Input received: [<tf.Tensor 'input_1:0' shape=(?, 100) dtype=float32>, <tf.Tensor 'input_2:0' shape=(?, 100) dtype=float32>]
解決方式：去掉參數 training=True
具體：bert_model = load_trained_model_from_checkpoint(config_path, checkpoint_path, output_layer_num=7, trainable=True, seq_len=Config.max_seq_len)
