What are the differences within the Google ALbert and modeling AlBert ?

[shuxiaobo]

Hello @ewrfcas , thanks for your contribution for this QA repo, it helps me a lot. I've reviewed the code and it works fine using my dataset, but here is a question about the code which in **[bert_cn_finetune.CHID_finetune.py file]**

bert_cn_finetune/CHID_finetune.py

Lines 241 to 250 in ec3cced

	if 'albert' in args.model_name:
	if 'google' in args.model_name:
	bert_config = AlbertConfig.from_json_file(args.bert_config_file)
	model = reset_model(args, bert_config, AlbertForMultipleChoice)
	else:
	bert_config = ALBertConfig.from_json_file(args.bert_config_file)
	model = reset_model(args, bert_config, ALBertForMultipleChoice)
	else:
	bert_config = BertConfig.from_json_file(args.bert_config_file)
	model = reset_model(args, bert_config, BertForMultipleChoice)

Could you help me to understand What are the differences within the Google ALbert and modeling AlBert ?

Thanks.