j-hartmann
commited on
Commit
•
adfea0d
1
Parent(s):
5e453b0
Update README.md
Browse files
README.md
CHANGED
@@ -93,6 +93,6 @@ Please find an overview of the datasets used for training below. All datasets co
|
|
93 |
|GoEmotions, Demszky et al. (2020)|Yes|Yes|Yes|Yes|Yes|Yes|Yes|
|
94 |
|ISEAR, Vikash (2018)|Yes|Yes|Yes|Yes|-|Yes|-|
|
95 |
|MELD, Poria et al. (2019)|Yes|Yes|Yes|Yes|Yes|Yes|Yes|
|
96 |
-
|SemEval-2018, EI-reg
|
97 |
|
98 |
The model is trained on a balanced subset from the datasets listed above (2,811 observations per emotion, i.e., nearly 20k observations in total). 80% of this balanced subset is used for training and 20% for evaluation. The evaluation accuracy is 66% (vs. the random-chance baseline of 1/7 = 14%).
|
|
|
93 |
|GoEmotions, Demszky et al. (2020)|Yes|Yes|Yes|Yes|Yes|Yes|Yes|
|
94 |
|ISEAR, Vikash (2018)|Yes|Yes|Yes|Yes|-|Yes|-|
|
95 |
|MELD, Poria et al. (2019)|Yes|Yes|Yes|Yes|Yes|Yes|Yes|
|
96 |
+
|SemEval-2018, EI-reg, Mohammad et al. (2018) |Yes|-|Yes|Yes|-|Yes|-|
|
97 |
|
98 |
The model is trained on a balanced subset from the datasets listed above (2,811 observations per emotion, i.e., nearly 20k observations in total). 80% of this balanced subset is used for training and 20% for evaluation. The evaluation accuracy is 66% (vs. the random-chance baseline of 1/7 = 14%).
|