Looker grid tag fixes for custom classifications

[benjaminpkane]

Consist rendering of classifications and other dynamic label fields in the App

Image

Dynamic documents

import fiftyone as fo
import fiftyone.zoo as foz

dataset = foz.load_zoo_dataset("quickstart", max_samples=10)

for sample in dataset:
    sample["dynamic"] = fo.DynamicEmbeddedDocument(
        classification=fo.Classification(label="single"),
        classifications=fo.Classifications(
            classifications=[fo.Classification(label="list")]
        ),
        raw_classifications=[fo.Classification(label="raw_list")],
        detection=fo.Detection(
            bounding_box=[0.25, 0.25, 0.5, 0.5], label="single"
        ),
        detections=fo.Detections(
            detections=[
                fo.Detection(bounding_box=[0.25, 0.25, 0.5, 0.5], label="list")
            ]
        ),
        raw_detections=[
            fo.Detection(bounding_box=[0.25, 0.25, 0.5, 0.5], label="raw_list")
        ],
    )

    sample.save()

dataset.add_dynamic_sample_fields()

session = fo.launch_app(dataset)

Lists of dynamic documents

import fiftyone as fo
import fiftyone.zoo as foz

dataset = foz.load_zoo_dataset("quickstart", max_samples=10)

for sample in dataset:
    sample["dynamic"] = [
        fo.DynamicEmbeddedDocument(
            annotator="alice",
            labels=fo.Classifications(
                classifications=[
                    fo.Classification(label="cat", mood="surly"),
                    fo.Classification(label="dog"),
                ]
            ),
        ),
        fo.DynamicEmbeddedDocument(
            annotator="bob",
            labels=fo.Classifications(
                classifications=[
                    fo.Classification(label="rabbit"),
                    fo.Classification(label="squirrel", age=51),
                ]
            ),
        ),
    ]
    sample.save()

dataset.add_dynamic_sample_fields()

session = fo.launch_app(dataset)

Video

import fiftyone as fo
import fiftyone.zoo as foz

dataset = foz.load_zoo_dataset("quickstart-video")
for sample in dataset:
    sample.frames[1]["dynamic"] = fo.DynamicEmbeddedDocument(
        classification=fo.Classification(label="single"),
        classifications=fo.Classifications(
            classifications=[fo.Classification(label="list")]
        ),
        raw_classifications=[fo.Classification(label="raw_list")],
        detection=fo.Detection(
            bounding_box=[0.25, 0.25, 0.5, 0.5], label="single"
        ),
        detections=fo.Detections(
            detections=[
                fo.Detection(bounding_box=[0.25, 0.25, 0.5, 0.5], label="list")
            ]
        ),
        raw_detections=[
            fo.Detection(bounding_box=[0.25, 0.25, 0.5, 0.5], label="raw_list")
        ],
    )

    sample.save()

dataset.add_dynamic_frame_fields()

session = fo.launch_app(dataset)

[benjaminpkane]

This work is incomplete, will follow up in a new PR