The organisers of EmoContext provided 3 different sets of data
- Train data (~30,000 examples)
- Dev data (~2,700 examples)
- Test data (~5,500 examples)
The links to individual datasets are avilable in the EmoContext group
Our code assumess 3 files in this folder with the nomenclature:
- train.txt
- dev.txt
- test.txt