Bengal Language Model

Bengali language model is build with fastai's ULMFit and ready for prediction and classfication task.

NB:

• This tool mostly followed inltk
• We separated Bengali part with better evaluation results

Installation

pip install bnlm

Evaluation Result

Language Model

• Accuracy 48.26% on validation dataset
• Perplexity: ~22.79

Training

To train with your own corpus follow this repository

Features and API

Download pretrained Model

To start, first download pretrained Language Model and Sentencepiece model

from bnlm.bnlm import download_models

download_models()

Predict N Words

from bnlm.bnlm import BengaliTokenizer
from bnlm.bnlm import predict_n_words
model_path = 'model'
input_sen = "আমি বাজারে"
output = predict_n_words(input_sen, 3, model_path)
print("Word Prediction: ", output)

Get Sentence Encoding

from bnlm.bnlm import BengaliTokenizer
from bnlm.bnlm import get_sentence_encoding
model_path = 'model'
sp_model = "model/bn_spm.model"
input_sentence = "আমি ভাত খাই।"
encoding = get_sentence_encoding(input_sentence, model_path, sp_model)
print("sentence encoding is: ", encoding)

Get Embedding Vectors

from bnlm.bnlm import BengaliTokenizer
from bnlm.bnlm import get_embedding_vectors
model_path = 'model'
sp_model = "model/bn_spm.model"
input_sentence = "আমি ভাত খাই।"
embed = get_embedding_vectors(input_sentence, model_path, sp_model)
print("sentence embedding is : ", embed)

Sentence Similarity

from bnlm.bnlm import BengaliTokenizer
from bnlm.bnlm import get_sentence_encoding
from bnlm.bnlm import get_similar_sentences
model_path = 'model'
sp_model = "model/bn_spm.model"
sentence_1 = "আমি ভাত খাই।"
sentence_2 = "আমি ভাত খাই।"
sim = get_sentence_similarity(sentence_1, sentence_2, model_path, sp_model)
print("similarity is: ", sim)

Get Simillar Sentences

from bnlm.bnlm import BengaliTokenizer
from bnlm.bnlm import get_embedding_vectors
from bnlm.bnlm import get_similar_sentences

model_path = 'model'
sp_model = "model/bn_spm.model"

input_sentence = "আমি ভাত খাই।"
sen_pred = get_similar_sentences(input_sentence, 3, model_path, sp_model)
print(sen_pred)

Classification

upcomming