No objectiveValue in the output json, after running a hyper parameters tuning job on Google's ML engine. .

Description

Hi, I'm new to t2t and I'm trying to follow this poetry generation example.

In the Hyperparameter tuning stage, after running:

%%bash
DATADIR=gs://${BUCKET}/poetry/data
OUTDIR=gs://${BUCKET}/poetry/model_hparam
JOBNAME=poetry_$(date -u +%y%m%d_%H%M%S)
echo $OUTDIR $REGION $JOBNAME
gsutil -m rm -rf $OUTDIR
echo "'Y'" | t2t-trainer \
  --data_dir=gs://${BUCKET}/poetry/subset \
  --t2t_usr_dir=./poetry/trainer \
  --problem=$PROBLEM \
  --model=transformer \
  --hparams_set=transformer_poetry \
  --output_dir=$OUTDIR \
  --hparams_range=transformer_poetry_range \
  --autotune_objective='metrics-poetry_line_problem/accuracy_per_sequence' \
  --autotune_maximize \
  --autotune_max_trials=4 \
  --autotune_parallel_trials=4 \
  --train_steps=7500 --cloud_mlengine --worker_gpu=8

I get the following output json:

{
  "completedTrialCount": "4",
  "trials": [
    {
      "trialId": "1",
      "hyperparameters": {
        "hp_attention_dropout": "0.68926619291305535",
        "hp_hidden_size": "256",
        "hp_learning_rate": "0.094830311097768627",
        "hp_num_hidden_layers": "3"
      }
    },
...
  ],
  "consumedMLUnits": 59.57,
  "isHyperparameterTuningJob": true
}

Whereas, I was expecting to find an objective value of the metric I've specified. In the notebook I'm following, it states that the output json should contain something like:

{
      "trialId": "37",
      "hyperparameters": {
        "hp_num_hidden_layers": "4",
        "hp_learning_rate": "0.026711152525921437",
        "hp_hidden_size": "512",
        "hp_attention_dropout": "0.60589466163419292"
      },
      "finalMetric": {
        "trainingStep": "8000",
        "objectiveValue": 0.0276162791997
      }

What have I missed? thanks...

Environment information

google cloud's data lab (I guess it is running on a docker) 

$ pip freeze | grep tensor
mesh-tensorflow==0.0.5
tensor2tensor==1.10.0
tensorboard==1.10.0
tensorflow==1.10.0

$ python -V
Python 2.7.15 :: Anaconda, Inc.

tensorflow / tensor2tensor

No objectiveValue in the output json, after running a hyper parameters tuning job on Google's ML engine. . #1527

Description

Environment information