spark xgboost4j how to print out train and validation loss in each round

I am using spark xgboost4j 0.80 with scala, wonder how to print out the train and validation loss in each round (for example set num_round as 100, print out train and validation loss in 100 rounds during training). I have made it before on the non distributed version xgboost like this:

While I don't figure out how to achieve this in xgboost4j, I found an example code while there's no changes in the stdout after I create the evaluation set and specify it in the XGBoostClassifier's parameters map. I have searched a lot on the Internet while don't find similar questions and solutions.

My code is like:

val params = Map("eta"       -> 0.1f,
        "lambda"      -> 0.1f,
        "max_depth"   -> 6,
        "objective"   -> "binary:logistic",
        "num_workers" -> 10
        "num_round"   -> 100,
        // "num_early_stopping_rounds" -> 10,
        // "maximize_evaluation_metrics" -> false,
        // "eval_metric" -> "logloss",
        "eval_sets" -> Map("eval" -> eval)
      )
val xgb = new XGBoostClassifier(params).
              setFeaturesCol("features").
              setLabelCol("label")

val model = xgb.fit(train)

Thanks in advance for your help.