Brandon Lee
Brandon Lee
I have correctly set up the dataset but I am seeing 0s for the counts what might be the cause of this? Command : `precise-test -t 0.85 trained_model.net dataset` ```...
for some wake words, we only get few dev/test samples as their transcript must be equal to wakeword For example, when I attempted to generate dataset for `love you`, dev...
create_raw_dataset.py takes quite a long time to generate datasets. I thinking multi-threading AudioDatasetMetadataWriter write will do the job. Also, this process terminates with segfault
Currently, the training process only uses positive dev/test sets for intermediate evaluation. We will first need to update the code to use some of negative sets before we can implement...
https://optuna.org/ might be a good option
some files under howl/data/dataset (labeller, phone) should be located outside as a common module that can be added separately I have noticed some classes are incorrectly located and can be...
Instead of searching for exact match, we can include singular or plural forms Maybe we can use some features of nltk https://www.nltk.org/index.html??
`token type = 'phone'` setting may not be working. we need to check its functionality and make sure an example is included
Trainer class that manages model training logics given a config
What is the reasoning behind GLIP and GroundingDINO not being present in OVOD (Table 1)? Is it because GLIP/GroundingDINO detects specific object referred by the text, While OVOD assumes that...