Text Classification
PyTorch
Safetensors
English
eurovoc
Inference Endpoints
scampion commited on
Commit
14bc646
1 Parent(s): 48f6bda

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -12
README.md CHANGED
@@ -6,18 +6,18 @@ language:
6
  - en
7
  metrics:
8
  - type: f1
9
- value: 0.XX
10
  name: micro F1
11
  args:
12
- threshold: 0.XX
13
  - type: NDCG@3
14
- value: 0.X
15
  name: NDCG@5
16
  - type: NDCG@5
17
- value: 0.XX
18
  name: NDCG@5
19
  - type: NDCG@10
20
- value: 0.XX
21
  name: NDCG@10
22
  tags:
23
  - eurovoc
@@ -53,14 +53,17 @@ model = EurovocTagger.from_pretrained("EuropeanParliament/eurovoc_eu")
53
  ## Metrics
54
 
55
 
56
- ### Eurlex57k Dataset
57
 
58
- | Metric | Value | Threshold Value |
59
- |------------|----------|-----------------|
60
- | Micro F1 | 0.XX | 0.XX |
61
- | NDCG@3 | 0.XX | - |
62
- | NDCG@5 | 0.XX | - |
63
- | NDCG@10 | 0.XX | - |
 
 
 
64
 
65
  These values are in line with the state of the art in the field, see the publication [Large Scale Legal Text Classification Using Transformer Models](https://arxiv.org/pdf/2010.12871.pdf).
66
 
 
6
  - en
7
  metrics:
8
  - type: f1
9
+ value: 0.8188
10
  name: micro F1
11
  args:
12
+ threshold: 0.41
13
  - type: NDCG@3
14
+ value: 0.8778
15
  name: NDCG@5
16
  - type: NDCG@5
17
+ value: 0.8625
18
  name: NDCG@5
19
  - type: NDCG@10
20
+ value: 0.8711
21
  name: NDCG@10
22
  tags:
23
  - eurovoc
 
53
  ## Metrics
54
 
55
 
56
+ ### Eurovoc Dataset version 23.08
57
 
58
+ Stratification ratio 90/10 for training/test and training/validation
59
+
60
+
61
+ | Metric | Value | Threshold Value |
62
+ |------------|------------|-----------------|
63
+ | Micro F1 | 0.8188 | 0.41 |
64
+ | NDCG@3 | 0.8778 | - |
65
+ | NDCG@5 | 0.8625 | - |
66
+ | NDCG@10 | 0.8711 | - |
67
 
68
  These values are in line with the state of the art in the field, see the publication [Large Scale Legal Text Classification Using Transformer Models](https://arxiv.org/pdf/2010.12871.pdf).
69