Add methods to initialize, annotate, and save ModelCards

[codesue]

The current process for annotating a model card without using the toolkit is to initialize an empty model card and manually assign values to each field. Sometimes this requires instantiating new lists and fields, and it's quite verbose.

Example from the standalone tutorial:

import model_card_toolkit as mct

model_card = mct.ModelCard()
model_card.model_details.name = 'Fine-tuned MobileNetV2 Model for Cats vs. Dogs'
model_card.model_details.overview = (
    'This model distinguishes cat and dog images. It uses the MobileNetV2 '
    'architecture (https://arxiv.org/abs/1801.04381) and is trained on the '
    'Cats vs Dogs dataset '
    '(https://www.tensorflow.org/datasets/catalog/cats_vs_dogs). This model '
    'performed with high accuracy on both Cat and Dog images.'
)
model_card.model_details.owners = [
    mct.Owner(name='Model Cards Team', contact='model-cards@example.com')
]
model_card.model_details.licenses = [mct.License(identifier='Apache-2.0')]

We can improve the process by adding methods to initialize, annotate, and save model cards, similarly to skops by @adrinjalali and model_card_creator by @paul-crease-ms.

import model_card_toolkit as mct

model_card = mct.ModelCard(
    'Fine-tuned MobileNetV2 Model for Cats vs. Dogs',
    overview=(
        'This model distinguishes cat and dog images. It uses the MobileNetV2 '
        'architecture (https://arxiv.org/abs/1801.04381) and is trained on the '
        'Cats vs Dogs dataset '
        '(https://www.tensorflow.org/datasets/catalog/cats_vs_dogs). This model '
        'performed with high accuracy on both Cat and Dog images.'
    ),
    **kwargs
)

model_card.add_owner(name='Model Cards Team', contact='model-cards@example.com')
model_card.add_license(identifier='Apache-2.0')
model_card.add_dataset(**kwargs)
model_card.add_metrics(**kwargs)
# etc.

model_card.render() 
model_card.save()
# etc.

[hanneshapke]

I love the idea! Makes the API very clean and straight forward.
Thank you @codesue.

[codesue]

So far we've implemented load, save, and render functions/methods.

import model_card_toolkit as mct

# load the model card
model_card = mct.load_model_card('model_cards/cats_vs_dogs.json')

# render the document for visual inspection
model_card_html = model_card.render()

# make some changes and save the updated model card
model_card.model_details.name = 'Updated Fine-tuned MobileNetV2 Model for Cats vs. Dogs'
model_card.save('model_cards/updated_cats_vs_dogs.json')

Other methods still need to be planned and implemented. These are up for grabs.