search

flairNLP / flair

A very simple framework for state-of-the-art Natural Language Processing (NLP)
https://flairnlp.github.io/flair/
Other
13.83k stars 2.09k forks source link

Train , Test , Validation cross validation split #1284

Closed OguzKircicek closed 4 years ago

OguzKircicek commented 4 years ago

A clear and concise description of what you want to know. Hi @alanakbik

I want to train , test and validation cross validation split my data but didn't see documentation. What a path ?

OguzKircicek commented 4 years ago

Ok. I found.

I analyzed source code.

train_length = len(train) dev_size: int = round(train_length / 10) splits = random_split(train, [train_length - dev_size, dev_size]) train = splits[0] dev = splits[1]

stale[bot] commented 4 years ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

myeghaneh commented 3 years ago

Ok. I found.

I analyzed source code.

train_length = len(train) dev_size: int = round(train_length / 10) splits = random_split(train, [train_length - dev_size, dev_size]) train = splits[0] dev = splits[1]

can you elaborate your response, I still do not know how to do cross-validation using flair ?

Astudnew commented 2 years ago

Ok. I found.

I analyzed the source code.

train_length = len(train) dev_size: int = round(train_length / 10) splits = random_split(train, [train_length - dev_size, dev_size]) train = splits[0] dev = splits[1]

Do you means that the cross-validation is used in Flair with k-fold 10? could you refer to that the part that used cross validation?