meaning of number of # characters in subwords?

commonsense / conceptnet-numberbatch

Other

1.29k stars 143 forks source link

meaning of number of # characters in subwords? #64

Open ugurcanozalp opened 2 years ago

ugurcanozalp commented 2 years ago

For continuation words. there are varying number of # signs. For example, in the 5 first words we have followings:

/c/de/####er
/c/de/###er
/c/de/##er

For example, if I have a word ending with "er", which one should I use?

Thanks..

frankier commented 2 years ago

As I understand it, Numberbatch does not have subword units, and you should prefer to use lemma forms when possible, since these are the forms in ConceptNet. If there are some odd entries like ####er, it could be because those have occurred exactly as-is in text.

RowenaHe commented 11 months ago

Hello, have you found out the reason for '#####'? I encountered the same problem while loading the downloaded and decompressed .txt.gz file. I tried to read the file with 'rb' but it didn't really help.

zohaibshahid969 commented 10 months ago

where i can find numberbatch-en-17.02.txt ??? someone please help me.

LostInDarkMath commented 10 months ago

where i can find numberbatch-en-17.02.txt ??? someone please help me.

https://conceptnet.s3.amazonaws.com/downloads/2017/numberbatch/numberbatch-17.02.txt.gz