Compute word frequencies for each word in captions. See code above for data structure
captions = [['#START#', 'a', 'man', 'holding', 'up', 'a', 'big', 'white', 'surfboard', '#END#'], ['#START#', 'a', 'surfer', 'holding', 'a', 'surf', 'board', ',', 'while', 'in', 'normal', 'cloths', '.', '#END#'], ['#START#', 'a', 'man', 'standing', 'next', 'to', 'a', 'white', 'surfboard', '.', '#END#'], ['#START#', 'a', 'man', 'that', 'is', 'standing', 'near', 'a', 'surfboard', '.', '#END#'], ['#START#', 'a', 'young', 'man', 'in', 'shorts', 'stands', 'in', 'a', 'room', 'with', 'a', 'body', 'board', '.', '#END#']], [['#START#', 'a', 'number', 'of', 'people', 'riding', 'bicycles', 'on', 'a', 'street', '#END#'], ['#START#', 'people', 'are', 'riding', 'bikes', 'down', 'a', 'street', 'with', 'a', 'tree', '.', '#END#'], ['#START#', 'some', 'people', 'sitting', 'on', 'bikes', 'by', 'the', 'sidewalk', '#END#'], ['#START#', 'people', 'riding', 'bikes', 'in', 'the', 'street', 'with', 'mopeds', 'parked', 'on', 'the', 'sidewalk', '.', '#END#'], ['#START#', 'two', 'adults', 'and', 'a', 'child', 'riding', 'bicycles', 'on', 'a', 'city', 'street', '#END#']], [['#START#', 'a', 'group', 'of', 'young', 'men', 'holding', 'skateboards', 'standing', 'in', 'front', 'of', 'a', 'store', '.', '#END#'], ['#START#', 'some', 'male', 'skateboarders', 'are', 'posing', 'for', 'a', 'picture', '#END#'], ['#START#', 'a', 'number', 'of', 'people', 'with', 'skate', 'boards', 'on', 'a', 'city', 'street', '#END#'], ['#START#', 'the', 'group', 'of', '7', 'men', 'are', 'hanging', 'out', 'with', 'skateboards', '.', '#END#'], ['#START#', 'group', 'of', 'skateboarder', 'posing', 'for', 'photo', 'outside', 'of', 'building', 'with', 'large', 'window', '.', '#END#']],
Step by step
Solved in 4 steps with 2 images
