what's the typical saving with FTZ file?

facebookresearch / fastText

Library for fast text representation and classification.

https://fasttext.cc/

MIT License

25.94k stars 4.72k forks source link

what's the typical saving with FTZ file? #821

Open prabhatM opened 5 years ago

prabhatM commented 5 years ago

Hi,

My original trained file (.bin) is 318 MB. After quantizing it reduced to 250 MB.

Is this much of reduction in the file size expected or there's a possibility of further reduction in size? I had read somewhere that the reduction is usually by 90%. Am I missing something?

Celebio commented 5 years ago

Hi @prabhatM , It highly depends on the quantization parameters. Could you copy/paste how you quantize your model so we can advise you?

Best regards, Onur