Commit
•
f60d104
1
Parent(s):
1570690
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,16 +1,15 @@
|
|
| 1 |
---
|
| 2 |
base_model: facebook/mbart-large-cc25
|
| 3 |
-
|
| 4 |
-
-
|
| 5 |
-
|
| 6 |
-
- name: es_nl+no_processing
|
| 7 |
-
results: []
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
| 11 |
-
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
| 14 |
|
| 15 |
This model is a fine-tuned version of [facebook/mbart-large-cc25](https://huggingface.co/facebook/mbart-large-cc25) on the None dataset.
|
| 16 |
It achieves the following results on the evaluation set:
|
|
|
|
| 1 |
---
|
| 2 |
base_model: facebook/mbart-large-cc25
|
| 3 |
+
language:
|
| 4 |
+
- nl
|
| 5 |
+
- es
|
|
|
|
|
|
|
| 6 |
---
|
| 7 |
|
| 8 |
+
# ES and NL to AMR parsing (stratified)
|
|
|
|
| 9 |
|
| 10 |
+
This version was trained on a subselection of the data. The AMR 3.0 corpus was translated to all the relevant languages. We then divided the dataset so
|
| 11 |
+
that in total we only see half of each language's dataset (so that in total we only see the full AMR 3.0 corpus in size once). In other words,
|
| 12 |
+
all languages were undersampled for research purposes.
|
| 13 |
|
| 14 |
This model is a fine-tuned version of [facebook/mbart-large-cc25](https://huggingface.co/facebook/mbart-large-cc25) on the None dataset.
|
| 15 |
It achieves the following results on the evaluation set:
|