Commit ·
c8dee7b
1
Parent(s): 3961ade
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Base model: [microsoft/DialoGPT-small](https://huggingface.co/microsoft/DialoGPT-small)
|
| 2 |
+
|
| 3 |
+
Fine tuned for dialogue response generation on the [Persuasion For Good Dataset](https://gitlab.com/ucdavisnlp/persuasionforgood) (Wang et al., 2019)
|
| 4 |
+
|
| 5 |
+
Three additional special tokens were added during the fine-tuning process:
|
| 6 |
+
- <|pad|> padding token
|
| 7 |
+
- <|user|> speaker control token to prompt user responses
|
| 8 |
+
- <|system|> speaker control token to prompt system responses
|
| 9 |
+
|
| 10 |
+
The following Dialogues were excluded:
|
| 11 |
+
- Those with donation amounts outside of the task range of [$0, $2].
|
| 12 |
+
- Those where a donation of 0 was made at the end of the task but a non-zero amount was pledged in the dialogue.
|
| 13 |
+
- Those with more than 800 words.
|
| 14 |
+
|
| 15 |
+
Stats:
|
| 16 |
+
- Training set: 519 dialogues
|
| 17 |
+
- Validation set: 58 dialogues
|
| 18 |
+
- ~20 utterances per dialogue
|