OpenNMT/OpenNMT-tf

Multi-decoder setup for multi-task learning

Open

#130 创建于 2018年5月22日

在 GitHub 查看
 (5 评论) (0 反应) (0 负责人)Python (1,428 star) (394 fork)batch import
enhancementhelp wanted

描述

Similarly to the ParallelEncoder, a ParallelDecoder setup could allow multi-task learning. This should not be too hard to implement but we need to take care of some details:

  • support separate values for the decoding parameters (beam_width, length_penalty, etc.),
  • parts of SequenceToSequence assume a single output head (e.g. loss computation, reverse vocabulary lookup, exported outputs for model serving, etc. which should be moved in the decoder itself)

贡献者指南

技术栈
pythontensorflow
领域
machine learningai
议题类型
feature
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
4
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
1-2 days
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
stale
清晰度议题是否清楚说明期望改动、验收标准和下一步。
mostly clear
前置要求
PythonTensorFlowsequence to sequence modelsOpenNMT tf codebase
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
30
研究方向
Examine the existing `ParallelEncoder` implementation for guidance. Focus on the `SequenceToSequence` class to identify all components that assume a single output head, such as loss computation and reverse vocabulary lookup. Review decoder parameter handling to understand how to support separate beam width and length penalty per decoder. Check linked issues and PRs for any prior discussion or partial implementations.