KnowBetterHelps
KnowBetterHelps
the utterance for nnet training usually contains only one sentence, so there won't be any points or commas in it, if have, should be delete before training. "encoding" means the...
_for some reason results are still 100% corr_ can you show me some example of you train/...../data.json? _how to load a pretrained model to further training_ I didn't find any...
looks like you were using the script directly for your own data. _рон не отрываясь смотрел на письмо которое уже начало с углов дымиться_ in chinese, one syllbale can be...
I am doing some similar work for code-switch recognition, which in english I gona using subword 'BPE' unit, not a letter. for a example, catch --> ca tch , not...
_can you say me, please, what files I should search for changes_ the script in data preparation, specically in generating data.json
it will create something like this, from exp/***/decode/data.json >hyn.trn 过 去 的 就 不 要 想 了 (T0055G2375-T0055G2375S0447) 天 气 下 降 注 意 身 体 (T0055G2286-T0055G2286S0457) 浦 中 市...
are you still in training train2.py? how do you train the NET1? I can not understand what the param 'case' is ...... can you help me to start the training...
Thank you for your kind help, I figured out the "case" just means one kind of hparams: "default.yaml" or "hparams.yaml" but after running command "python train1.py -case "train1_net" -gpu 0"...
I have successfully running the NET1 training, Thanks for you kind help!
does it python3 supported now?