anandi1989/10k-word-embeddings

Word embeddings learned from 10-K documents

10K-word-embeddings

Word embeddings learned from 10-K documents as decribed in paper:
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3480902

Getting Started

These instructions will help you get started with using the word embeddings in your projects.

Prerequisities

Install PyTorch and Numpy

Details

10k_word_embeddings.tar - Tar file containing word embeddings. Each embedding is of 300 dimension.
vocab_to_int.tar - Tar file containing vocab to integer mapping. Vocab size is 159647 words.

Usage

import torch
embed = torch.load('10k_word_embeddings.tar')
vocab_to_int = torch.load('vocab_to_int.tar')

• Use learned embeddings as pre-trained embeddings in a neural network:
  from torch import nn
  embeddings = nn.Embedding(embed.shape[0], embed.shape[1])
  embeddings.weight.data.copy_(torch.from_numpy(embed))
  embeddings.weight.requires_grad = False

• Get learned embeddings of a word from vocabulary:
  embeddings_profit = embed[vocab_to_int['profit']]