Update README.md
Browse files
README.md
CHANGED
|
@@ -7,7 +7,6 @@ tags:
|
|
| 7 |
- doclaynet
|
| 8 |
- e5
|
| 9 |
pipeline_tag: text-classification
|
| 10 |
-
license: mit
|
| 11 |
base_model: intfloat/multilingual-e5-large
|
| 12 |
datasets:
|
| 13 |
- pierreguillou/DocLayNet-base
|
|
@@ -23,21 +22,22 @@ model-index:
|
|
| 23 |
name: DocLayNet
|
| 24 |
type: pierreguillou/DocLayNet-base
|
| 25 |
metrics:
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
|
|
|
|
| 41 |
---
|
| 42 |
# Multilingual E5 for Document Classification (DocLayNet)
|
| 43 |
This model is a fine-tuned version of intfloat/multilingual-e5-large for document text classification based on the DocLayNet dataset.
|
|
@@ -115,4 +115,4 @@ label_mapping = {
|
|
| 115 |
}
|
| 116 |
predicted_label = label_mapping[predicted_class]
|
| 117 |
|
| 118 |
-
```
|
|
|
|
| 7 |
- doclaynet
|
| 8 |
- e5
|
| 9 |
pipeline_tag: text-classification
|
|
|
|
| 10 |
base_model: intfloat/multilingual-e5-large
|
| 11 |
datasets:
|
| 12 |
- pierreguillou/DocLayNet-base
|
|
|
|
| 22 |
name: DocLayNet
|
| 23 |
type: pierreguillou/DocLayNet-base
|
| 24 |
metrics:
|
| 25 |
+
- type: accuracy
|
| 26 |
+
value: 0.9719
|
| 27 |
+
name: Test Accuracy
|
| 28 |
+
- type: f1
|
| 29 |
+
value: 0.972
|
| 30 |
+
name: Weighted F1 Score
|
| 31 |
+
- type: precision
|
| 32 |
+
value: 0.9732
|
| 33 |
+
name: Weighted Precision
|
| 34 |
+
- type: recall
|
| 35 |
+
value: 0.9719
|
| 36 |
+
name: Weighted Recall
|
| 37 |
+
- type: loss
|
| 38 |
+
value: 0.5192
|
| 39 |
+
name: Test Loss
|
| 40 |
+
library_name: transformers
|
| 41 |
---
|
| 42 |
# Multilingual E5 for Document Classification (DocLayNet)
|
| 43 |
This model is a fine-tuned version of intfloat/multilingual-e5-large for document text classification based on the DocLayNet dataset.
|
|
|
|
| 115 |
}
|
| 116 |
predicted_label = label_mapping[predicted_class]
|
| 117 |
|
| 118 |
+
```
|