yangperasd/bert_serving repository overview

⁠description

Use bert with sentence_piece as a web service to encode sentence.

Note: You should download a bert model with senence_piece on your computer.

⁠Usage:

• On server side, run this container.
  • cmd: docker run -v $(pwd)/model/:/model -p 5555:5555 -p 5556:5556 yangperasd/bert_serving 4 512
  • comment: docker run -v bert_model_dir:/model -p data_transfer_to_server_port:5555 -p data_transfer_to_client_port:5556 yangperasd/bert_serving num_worker max_sentence_length
• On client side, use bert-serving-client to get encoded sentence. code example: from bert_serving.client import BertClient bc = BertClient() bc.encode(['hello word'])

⁠Reference:

1. https://github.com/hanxiao/bert-as-service⁠
2. https://github.com/yoheikikuta/bert-japanese⁠