Is there a preferred way to do this?
It’s easy enough to build a vocab using the Tokenizer class in the .text namespace and some python … but including things like min/max_freq, etc… that come out-of-the-box with torchtext aren’t there.
Thanks - wg
Is there a preferred way to do this?
It’s easy enough to build a vocab using the Tokenizer class in the .text namespace and some python … but including things like min/max_freq, etc… that come out-of-the-box with torchtext aren’t there.
Thanks - wg