One-Stop Solution to encode sentence to fixed length vectors from various embedding techniques
• Inspired from bert-as-service

What is it • Installation • Getting Started • Supported Embeddings • API •

Encoding/Embedding is a upstream task of encoding any inputs in the form of text, image, audio, video, transactional data to fixed length vector. Embeddings are quite popular in the field of NLP, there has been various Embeddings models being proposed in recent years by researchers, some of the famous one are bert, xlnet, word2vec etc. The goal of this repo is to build one stop solution for all embeddings techniques available, here we are starting with popular text embeddings for now and later on we aim to add as much technique for image, audio, video inputs also.

embedding-as-service help you to encode any given text to fixed length vector from supported embeddings and models.

^{▴ Back to top}

Here we have given the capability to use embedding-as-service like a module or you can run it as a server and handle queries by installing client package embedding-as-service-client

Install the embedding-as-servive via pip.

$ pip install embedding-as-service

Note that the code MUST be running on Python >= 3.6. Again module does not support Python 2!

Here you also need to install a client module embedding-as-service-client

$ pip install embedding-as-service # server
$ pip install embedding-as-service-client # client

Client module need not to be on Python 3.6, it supports both Python2 and Python3

^{▴ Back to top}

If using embedding-as-service as a module

>>> from embedding_as_service.text.encode import Encoder  
>>> en = Encoder(embedding='bert', model='bert_base_cased', max_seq_length=256)  

If using embedding-as-service as a server

# start the server by proving embedding, model, port, max_seq_length[default=256], num_workers[default=4]
$ embedding-as-service-start --embedding bert --model bert_base_cased --port 8080 --max_seq_length 256

>>> from embedding_as_service_client import EmbeddingClient
>>> en = EmbeddingClient(host=<host_server_ip>, port=<host_port>)

>>> vecs = en.encode(texts=['hello aman', 'how are you?'])  
>>> vecs  
array([[[ 1.7049843 ,  0.        ,  1.3486509 , ..., -1.3647075 ,  
 0.6958289 ,  1.8013777 ], ... [ 0.4913215 ,  0.60877025,  0.73050433, ..., -0.64490885, 0.8525057 ,  0.3080206 ]]], dtype=float32)  
>>> vecs.shape  
(2, 128, 768) # batch x max_sequence_length x embedding_size  

>>> vecs = en.encode(texts=['hello aman', 'how are you?'], pooling='reduce_mean')  
>>> vecs  
array([[-0.33547154,  0.34566957,  1.1954105 , ...,  0.33702594,  
 1.0317835 , -0.785943  ], [-0.3439088 ,  0.36881036,  1.0612687 , ...,  0.28851607, 1.1107115 , -0.6253736 ]], dtype=float32)  
  
>>> vecs.shape  
(2, 768) # batch x embedding_size  

>>> en.tokenize(texts=['hello aman', 'how are you?'])  
[['_hello', '_aman'], ['_how', '_are', '_you', '?']]  

>>> texts = ['hello aman!', 'how are you']  
  
# a naive whitespace tokenizer  
>>> tokens = [s.split() for s in texts]  
>>> vecs = en.encode(tokens, is_tokenized=True)  

^{▴ Back to top}

1. class embedding_as_service.text.encoder.Encoder

2. def embedding_as_service.text.encoder.Encoder.encode

3. def embedding_as_service.text.encoder.Encoder.tokenize

^{▴ Back to top}

Here are the list of supported embeddings and their respective models.

This software uses the following open source packages:

• XLnet
• tensorflow-hub

Thanks goes to these wonderful people (emoji key):

MrPranav101 💻 📖 🚇

Aman Srivastava 💻 📖 🚇

Chirag Jain 💻 📖 🚇

Ashutosh Singh 💻 📖 🚇

Dhaval Taunk 💻 📖 🚇

Alec Koumjian 🐛

Pradeesh 🐛

This project follows the all-contributors specification. Contributions of any kind welcome!

Please read the contribution guidelines first.

^{▴ Back to top}

If you use embedding-as-service in a scientific publication, we would appreciate references to the following BibTex entry:

@misc{aman2019embeddingservice,
  title={embedding-as-service},
  author={Srivastava, Aman},
  howpublished={\url{https://github.com/amansrivastava17/embedding-as-service}},
  year={2019}
}