Commit
•
5d851a5
1
Parent(s):
c0107d1
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,19 +1,87 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
| 5 |
model-index:
|
| 6 |
-
- name: verdict-classifier
|
| 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 [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the None dataset.
|
| 16 |
-
It achieves the following results on the evaluation set:
|
| 17 |
- Loss: 0.2238
|
| 18 |
- F1 Macro: 0.8540
|
| 19 |
- F1 Misinformation: 0.9798
|
|
@@ -24,17 +92,6 @@ It achieves the following results on the evaluation set:
|
|
| 24 |
- Prec Factual: 0.9889
|
| 25 |
- Prec Other: 0.5294
|
| 26 |
|
| 27 |
-
## Model description
|
| 28 |
-
|
| 29 |
-
More information needed
|
| 30 |
-
|
| 31 |
-
## Intended uses & limitations
|
| 32 |
-
|
| 33 |
-
More information needed
|
| 34 |
-
|
| 35 |
-
## Training and evaluation data
|
| 36 |
-
|
| 37 |
-
More information needed
|
| 38 |
|
| 39 |
## Training procedure
|
| 40 |
|
|
@@ -102,4 +159,4 @@ The following hyperparameters were used during training:
|
|
| 102 |
- Transformers 4.11.3
|
| 103 |
- Pytorch 1.9.0+cu102
|
| 104 |
- Datasets 1.9.0
|
| 105 |
-
- Tokenizers 0.10.2
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- am
|
| 5 |
+
- ar
|
| 6 |
+
- hy
|
| 7 |
+
- eu
|
| 8 |
+
- bn
|
| 9 |
+
- bs
|
| 10 |
+
- bg
|
| 11 |
+
- my
|
| 12 |
+
- hr
|
| 13 |
+
- ca
|
| 14 |
+
- cs
|
| 15 |
+
- da
|
| 16 |
+
- nl
|
| 17 |
+
- en
|
| 18 |
+
- et
|
| 19 |
+
- fi
|
| 20 |
+
- fr
|
| 21 |
+
- ka
|
| 22 |
+
- de
|
| 23 |
+
- el
|
| 24 |
+
- gu
|
| 25 |
+
- ht
|
| 26 |
+
- iw
|
| 27 |
+
- hi
|
| 28 |
+
- hu
|
| 29 |
+
- is
|
| 30 |
+
- in
|
| 31 |
+
- it
|
| 32 |
+
- ja
|
| 33 |
+
- kn
|
| 34 |
+
- km
|
| 35 |
+
- ko
|
| 36 |
+
- lo
|
| 37 |
+
- lv
|
| 38 |
+
- lt
|
| 39 |
+
- ml
|
| 40 |
+
- mr
|
| 41 |
+
- ne
|
| 42 |
+
- no
|
| 43 |
+
- or
|
| 44 |
+
- pa
|
| 45 |
+
- ps
|
| 46 |
+
- fa
|
| 47 |
+
- pl
|
| 48 |
+
- pt
|
| 49 |
+
- ro
|
| 50 |
+
- ru
|
| 51 |
+
- sr
|
| 52 |
+
- zh
|
| 53 |
+
- sd
|
| 54 |
+
- si
|
| 55 |
+
- sk
|
| 56 |
+
- sl
|
| 57 |
+
- es
|
| 58 |
+
- sv
|
| 59 |
+
- tl
|
| 60 |
+
- ta
|
| 61 |
+
- te
|
| 62 |
+
- th
|
| 63 |
+
- tr
|
| 64 |
+
- uk
|
| 65 |
+
- ur
|
| 66 |
+
- ug
|
| 67 |
+
- vi
|
| 68 |
+
- cy
|
| 69 |
tags:
|
| 70 |
- generated_from_trainer
|
| 71 |
model-index:
|
| 72 |
+
- name: verdict-classifier-en
|
| 73 |
+
results:
|
| 74 |
+
- task:
|
| 75 |
+
type: text-classification
|
| 76 |
+
name: Verdict Classification
|
| 77 |
+
widget:
|
| 78 |
+
- "Even though it might look true, it has been taken out of context."
|
| 79 |
---
|
| 80 |
|
| 81 |
+
# Multilingual Verdict Classifier
|
|
|
|
| 82 |
|
| 83 |
+
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on 2,500 deduplicated multilingual verdicts from [Google Fact Check Tools API](https://developers.google.com/fact-check/tools/api/reference/rest/v1alpha1/claims/search), translated into 65 languages with the [Google Cloud Translation API](https://cloud.google.com/translate/docs/reference/rest/).
|
| 84 |
+
It achieves the following results on the evaluation set, being 1,000 such verdicts, but here including duplicates to represent the true distribution:
|
|
|
|
|
|
|
| 85 |
- Loss: 0.2238
|
| 86 |
- F1 Macro: 0.8540
|
| 87 |
- F1 Misinformation: 0.9798
|
|
|
|
| 92 |
- Prec Factual: 0.9889
|
| 93 |
- Prec Other: 0.5294
|
| 94 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 95 |
|
| 96 |
## Training procedure
|
| 97 |
|
|
|
|
| 159 |
- Transformers 4.11.3
|
| 160 |
- Pytorch 1.9.0+cu102
|
| 161 |
- Datasets 1.9.0
|
| 162 |
+
- Tokenizers 0.10.2
|