DeepQA
DeepQA copied to clipboard
why if after train i chage settings in params.ini training crashes and lost ?
do i somthin misunderstund ??
- trained with 4 words max lengt to 0.7
- open params ini in save dir changed seting to max 6 run train and it ruin all no way to recover
From what I remember changing the max length will change the vocabulary size used. So the program will crash because the vocabulary embedding shape does not match. I think it should be possible to change the max length if you make sure the same vocabulary is used.
but is there solution ?? because i want to inject in training new conversations and voc will grow anyway ...
same to tensor size my plan is to incarase from 6 words in sentense atm to more. but how to keep previous indexses ?? or some clever way to rebuild it ... ??
not only crash ... ya lost previous training data :D
and next i want to split by sylabies all words .... in far future when i get some nvidia gfx card with ton of cudcores