khulnasoft/bangla_word2vec

Bengali Word2Vec Model

This is a pre-trained word2vec model for Bengali language.

This model is build for bengalinlp package.

Datasets

• Wikipedia dump datasets

Training details

• Word2Vec word embedding dimension = 100, min_count=5, window=5, epochs=10

Usage

• pip install -U bengalinlp_toolkit

• Generate Vector using pretrain model

  from bengalinlp import BengaliWord2Vec
  
  bwv = BengaliWord2Vec()
  model_path = "bengali_word2vec.model"
  word = 'গ্রাম'
  vector = bwv.generate_word_vector(model_path, word)
  print(vector.shape)
  print(vector)
  

• Find Most Similar Word Using Pretrained Model

  from bengalinlp import BengaliWord2Vec
  
  bwv = BengaliWord2Vec()
  model_path = "bengali_word2vec.model"
  word = 'গ্রাম'
  similar = bwv.most_similar(model_path, word, topn=10)
  print(similar)