diff options
author | TharinduDR <rhtdranasinghe@gmail.com> | 2020-10-13 21:40:02 +0300 |
---|---|---|
committer | TharinduDR <rhtdranasinghe@gmail.com> | 2020-10-13 21:40:02 +0300 |
commit | e9499a6fcccaf5ccdbddf7b1180c476f5d80a167 (patch) | |
tree | 704a66cc6897da47315acb1e8f5a2f04a4833dac /docs | |
parent | 78c425a3b0545ab69f5078ee28958d96c562839c (diff) |
033: Adding documentation
Diffstat (limited to 'docs')
-rw-r--r-- | docs/examples.md | 21 |
1 files changed, 21 insertions, 0 deletions
diff --git a/docs/examples.md b/docs/examples.md index 9c0c388..03689a6 100644 --- a/docs/examples.md +++ b/docs/examples.md @@ -56,3 +56,24 @@ Both architectures in TransQuest outperforms OpenKiwi in all the language pairs. | | OpenKiwi | 0.1676 | 0.6559 | 0.8503 | +## [WMT 2020 QE Task 2: Sentence-Level Post-editing Effort](http://www.statmt.org/wmt20/quality-estimation-task.html) +This task consists predicting Sentence-level HTER (Human Translation Error Rate) scores for a given source and a target. + +To run the experiments for each language please run this command from the root directory of TransQuest. + +```bash +python -m examples.wmt_2020_task2.<language-pair>.<architecture> +``` + +Language Pair options : en_zh (English-Chinese), en_de (English-German) + +Architecture Options : trans_quest (MonoTransQuest), siamese_trans_quest (SiameseTransQuest). + +As an example to run the experiments on English-Chinese with MonoTransQuest architecture, run the following command. + +```bash +python -m examples.wmt_2020_task2.en_zh.trans_quest +``` + +### Results +Both architectures in TransQuest outperforms OpenKiwi in all the language pairs. Furthermore, TransQuest won this task in all the language pairs. |