asahi417 commited on
Commit
cb81707
1 Parent(s): ab58ad6

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -88,6 +88,7 @@ print(target_format)
88
 
89
  For the temporal-shift setting, model should be trained on `train_2020` with `validation_2020` and evaluate on `test_2021`.
90
  In general, model would be trained on `train_all`, the most representative training set with `validation_2021` and evaluate on `test_2021`.
 
91
  **IMPORTANT NOTE:** To get a result that is comparable with the results of the COLING 2022 Tweet Topic paper, please use `train_coling2022` and `test_coling2022` for temporal-shift, and `train_coling2022_random` and `test_coling2022_random` fir random split (the coling2022 split does not have validation set).
92
 
93
  ### Models
88
 
89
  For the temporal-shift setting, model should be trained on `train_2020` with `validation_2020` and evaluate on `test_2021`.
90
  In general, model would be trained on `train_all`, the most representative training set with `validation_2021` and evaluate on `test_2021`.
91
+
92
  **IMPORTANT NOTE:** To get a result that is comparable with the results of the COLING 2022 Tweet Topic paper, please use `train_coling2022` and `test_coling2022` for temporal-shift, and `train_coling2022_random` and `test_coling2022_random` fir random split (the coling2022 split does not have validation set).
93
 
94
  ### Models