Test data intruncated

[KiddoKiddo]

Hi,
I have an issue that my test data has total 36790 rows, but after using the prediction.py, my output file only has 36560 rows. I already checked dev_parsed.json, which has enough data. Can you help to pin point what is the problem? Many thanks.
Regards,
Thy

[EireneX]

Encountered same issue. Could it be related to word2vec?
Saw some WARN message in log during proprocess:

 WARN edu.stanford.nlp.process.PTBLexer - Untokenizable: ‍ (U+200D, decimal: 8205)
[main] WARN edu.stanford.nlp.process.PTBLexer - Untokenizable: ‍ (U+200D, decimal: 8205)
[main] WARN edu.stanford.nlp.process.PTBLexer - Untokenizable: ‍ (U+200D, decimal: 8205)
[main] WARN edu.stanford.nlp.process.PTBLexer - Untokenizable: ‍ (U+200D, decimal: 8205)
[main] WARN edu.stanford.nlp.process.PTBLexer - Untokenizable: ‍ (U+200D, decimal: 8205)
[main] WARN edu.stanford.nlp.process.PTBLexer - Untokenizable: ‍ (U+200D, decimal: 8205)
[main] WARN edu.stanford.nlp.process.PTBLexer - Untokenizable: ‍ (U+200D, decimal: 8205)
[main] WARN edu.stanford.nlp.process.PTBLexer - Untokenizable: ‍ (U+200D, decimal: 8205)
[main] WARN edu.stanford.nlp.process.PTBLexer - Untokenizable: ‍ (U+200D, decimal: 8205)
[main] WARN edu.stanford.nlp.process.PTBLexer - Untokenizable: ‍ (U+200D, decimal: 8205)

[YouseYeung]

Hi,
I also want to use this model to test my dataset. But I got an error. Can you tell me how to define my own dataset in a file?