PyTorch学习笔记之初识word_embedding
Posted Joyce_song94
tags:
篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了PyTorch学习笔记之初识word_embedding相关的知识,希望对你有一定的参考价值。
1 import torch 2 import torch.nn as nn 3 from torch.autograd import Variable 4 5 word2id = {‘hello‘: 0, ‘world‘: 1} 6 # you have 2 words, and then need 5 dim each word 7 embeds = nn.Embedding(2, 5) 8 # we need variable, because we need use element of nn.Embedding 9 hello_idx = torch.LongTensor([word2id[‘hello‘]]) 10 print(hello_idx) 11 hello_idx = Variable(hello_idx) 12 print(hello_idx) 13 # achieve the initial word_embedding of ‘hello‘ 14 hello_embed = embeds(hello_idx) 15 print(hello_embed) 16 ‘‘‘ 17 0 18 [torch.LongTensor of size 1] 19 20 Variable containing: 21 0 22 [torch.LongTensor of size 1] 23 24 Variable containing: 25 1.1842 0.6819 -0.8768 -1.5130 0.7650 26 [torch.FloatTensor of size 1x5] 27 ‘‘‘
以上是关于PyTorch学习笔记之初识word_embedding的主要内容,如果未能解决你的问题,请参考以下文章