Having problems running on fasttext vectors

[patrickdrouin]

Hello,

I am trying to test meemi on some fasttext embeddings (in txt format) I have. Here's the output I get. Any idea what is happening? I suspect it can read the embeddings since the first 2 steps seem to be running just fine.

(metavirt) $ ./meemi/get_crossembs.sh corpus_en.tag.bin.vec corpus_fr.tag.bin.vec en-fr.txt
Cross-lingual means
Mean mapping
Traceback (most recent call last):
File "map_embeddings.py", line 304, in
main()
File "map_embeddings.py", line 83, in main
trg_words, z = embeddings.read(trgfile, dtype=dtype)
File ".../metavirt/meemi/utils/vecmap/embeddings.py", line 23, in read
count = int(header[0]) if threshold <= 0 else min(threshold, int(header[0]))
ValueError: invalid literal for int() with base 10: ''
Traceback (most recent call last):
File "map_embeddings.py", line 304, in
main()
File "map_embeddings.py", line 83, in main
trg_words, z = embeddings.read(trgfile, dtype=dtype)
File ".../metavirt/meemi/utils/vecmap/embeddings.py", line 23, in read
count = int(header[0]) if threshold <= 0 else min(threshold, int(header[0]))
ValueError: invalid literal for int() with base 10: ''

[patrickdrouin]

Tested with word2vec embeddings and ran into the same problem.