[Object Detection][Feature Request] Allow more than one eval_spec

• What is the top-level directory of the model you are using: object_detection
• Have I written custom code: no
• OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Ubuntu 16.04
• TensorFlow installed from (source or binary): binary
• TensorFlow version (use command below): 1.10.
• Bazel version (if compiling from source): -
• CUDA/cuDNN version: CUDA 9
• GPU model and memory: 1080Ti, 11GB
• Exact command to reproduce: Evaluating on more than one dataset/metrics_set while training

Describe the problem

In training and evaluation case of model_main you restrict to a single evaluation spec.
It would be very appreciated if could support more than one evaluation spec.
This would mean another evaluation sets (e.g. validation sets of several datasets), with corresponding protocols (metrics_set).

Thanks in advance.