import datasets | |
db = datasets.load_dataset('vivym/midjourney-prompts', cache_dir='./tmp') | |
captions = list(set(db['train']['prompt'])) | |
print(len(captions)) | |
dataset = datasets.Dataset.from_dict({"text": captions}) | |
dataset.push_to_hub("Geonmo/midjourney-prompts-only") | |