cerebras.modelzoo.data.multimodal.datasets.features.VQAFeaturesDict#

class cerebras.modelzoo.data.multimodal.datasets.features.VQAFeaturesDict(image_path, image_id, question, answers, multiple_choice_answer, multiple_choice_answer_language, answer_type=None, image=None)[source]#

Bases: object

ImageID: Name of image

Methods

Attributes

answer_type

image

image_path

image_id

question

answers

multiple_choice_answer

multiple_choice_answer_language

previous

cerebras.modelzoo.data.multimodal.datasets.features.VQAAnswer

next

cerebras.modelzoo.data.multimodal.datasets.features.VQAQuestion