Lighten the datasets - Githubissues

ngarneau / vecmap

A framework to learn cross-lingual word embedding mappings

GNU General Public License v3.0

2 stars 1 forks source link

Closed ngarneau closed 5 years ago

ngarneau commented 5 years ago

There is 2-3 ish different datasets:

the dataset of Zhang et al. (2017a)
the dataset of Dinu et al. (2015)
the extension of Din et al. (2015) of Artetxe et al. (2017, 2018a) Make sure to know which datasets we have in the repo and if we need to gather some at other places. This need to be documented in our paper (the fuzziness of which dataset is used in the paper vs code)

ngarneau commented 5 years ago

Found out that we have the dataset from Dinu et al and the extension. We don't have the dataset of Zhang to reproduce the results of Table 1