Is BertTokenizer similar to word embedding?

Question

The idea of using BertTokenizer from huggingface really confuses me.

1. When I use

   tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
   tokenizer.encode_plus("Hello")
   

Does the result is somewhat similar to when I pass a one-hot vector representing "Hello" to a learning embedding matrix?

2. How is

   BertTokenizer.from_pretrained("bert-base-uncased") 
   

different from

BertTokenizer.from_pretrained("bert-**large**-uncased") 

and other pretrained?

Answer 1

The encode_plus and encode functions tokenize your texts and prepare them in a proper input format of the BERT model. Therefore you can see them similar to the one-hot vector in your provided example.
The encode_plus returns a BatchEncoding consisting of input_ids, token_type_ids, and attention_mask.

The pre-trained model differs based on the number of encoder layers. The base model has 12 encoders, and the large model has 24 layers of encoders.