Skip to content

Commit

Permalink
update readme
Browse files Browse the repository at this point in the history
  • Loading branch information
longdt219 committed Aug 24, 2016
1 parent 48e5c50 commit 4c9a983
Showing 1 changed file with 5 additions and 3 deletions.
8 changes: 5 additions & 3 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ If you use this code, please cite the paper
publisher = {Association for Computational Linguistics},
}
```
#### Getting started
### Getting started
The implementation is basically the extension of C version Word2Vec. You just need to do the make
make

Expand All @@ -26,11 +26,13 @@ We also included a tiny mixed English-Italian monolingual data for demo purposes
/data/mono/en_it.shuf.10k
The following will build the crosslingual word embeddings for English and Italian.
```
./xlingemb -train data/mono/en_it.shuf.10k -output en.it.word.emb -size 200 -window 48 -iter 15 -negative 25 -sample 0.0001 -alpha 0.025 -cbow 1 -threads 5 -dict data/dicts/en.it.panlex.all.processed -outputn en.it.context.emb -reg 0.01
./xlingemb -train data/mono/en_it.shuf.10k -output en.it.word.emb -size 200 -window 48 -iter 15
-negative 25 -sample 0.0001 -alpha 0.025 -cbow 1 -threads 5
-dict data/dicts/en.it.panlex.all.processed -outputn en.it.context.emb -reg 0.01
```
Some options :
- output: the usual word embeddings
- outputn : the context word embeddings which is the final output.
- <b>outputn</b> : the context word embeddings which is the final output.
- size, window, iter, negative, sample, alpha, cbow, threads : the same as Word2Vec
- dict: the dictionary
- reg: the regulariser sensitivity for combining word and context embeddings.
Expand Down

0 comments on commit 4c9a983

Please sign in to comment.