Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
Modules can hold parameters of different types on different devices, so it's not always possible to unambiguously determine the device. The recommended workflow in PyTorch is to create the device object …
Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
nltk.util.ngrams nltk.bigrams nltk.PorterStemmer
from nltk.util import ngrams
sentence = 'this is a foo bar sentences and i want to ngramize it'
n = 6
sixgrams = ngrams(sentence.split …Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
existing stemming method such as NLTK.PorterStem, etc.
didn't -> did not, there's -> there is, etc. Mr. -> Mister Mrs. -> ... Ms. -> ...
it seems that it is hard to get …
Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
PyTorch uses pickle to serialize and deserialize objects.
The PyTorch convention is
to use the file extension .pt or .pth for saving model (or its parameters)
and use the file extension …
Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
https://github.com/dmlc/dgl
Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!
https://en.wikipedia.org/wiki/Cosine_similarity
https://en.wikipedia.org/wiki/Jaccard_index