Feature Request

I want to obtain the evaluation of different datasets of a single job
lumi eval -h && lumi eval -c config_ssd.yml --split valid --split train --split test --watch --from-global-step 0

I understand that it is not so trivial because it was being kept as a summary of work scalars like this
![image](https://user-images.githubusercontent.com/7362688/48442210-8be6d080-e75b

Currently, it seems impossible to just get a list of bounding boxes from the predictions.

img = np.array(Image.open("..."))
infer_ds = Dataset.from_images([img], valid_tfms, class_map=dm.parser.class_map)

preds = model_type.predict(model, infer_ds, keep_images=True)
# I just want the bounding boxes 

preds is a list of Prediction and the codebase is not easy to navi