Contrastive Language Image Pretrain

数据集目录格式
└─Flickr8k
   ├─Flickr8k_Dataset
   |      └─Flickr8k_Dataset
   └─Flickr8k_text
          ├─Flickr8k.devImages.txt
          ├─Flickr8k.testImages.txt
          ├─Flickr8k.trainImages.txt
          └─Flickr8k.token.txt

import mindspore; mindspore.set_context(mode=0, device_id=0)
from mindformers import MindFormerBook
from mindformers.trainer import Trainer

# 显示Trainer的模型支持列表
MindFormerBook.show_trainer_support_model_list("contrastive_language_image_pretrain")
# INFO - Trainer support model list for contrastive_language_image_pretrain task is:
# INFO -    ['clip_vit_b_32', 'clip_vit_b_16', 'clip_vit_l_14', 'clip_vit_l_14@336']
# INFO - -------------------------------------

# 初始化trainer
trainer = Trainer(task='contrastive_language_image_pretrain',
    model='clip_vit_b_32',
    train_dataset='./Flickr8k'
)

trainer.train()