什么是tf.contrib.seq2seq.TrainingHelper中的sequence_length字段？

Question

任何人都可以在class BooksSpider(scrapy.Spider): name = 'books' allowed_domains = ['books.toscrape.com'] start_urls = ['http://books.toscrape.com/'] allTitles = [] def parse(self, response): books = response.xpath('//h3/a/@href').extract() for book in books: absolute_url = response.urljoin(book) yield scrapy.Request(absolute_url, callback=self.parse_book) # process next page next_page_url = response.xpath('//a[text()="next"]/@href').extract_first() absolute_next_page_url = response.urljoin(next_page_url) yield scrapy.Request(absolute_next_page_url) def parse_book(self, response): title = response.css('h1::text').extract_first() self.allTitles.extend([ title ]) price = response.xpath('//*[@class="price_color"]/text()').extract_first() yield {'title':title, 'price': price}

中解释序列长度字段的用途

Answer 1

它代表“输入”参数中每个目标序列的长度。

检查：例如http://higepon.hatenablog.com/entry/20171212/1513076578。

用于在将每个令牌馈送到解码器时检查目标序列令牌是否已用尽。检查以下内容： https://github.com/tensorflow/tensorflow/blob/r1.10/tensorflow/contrib/seq2seq/python/ops/helper.py#L244