z-dickson commited on
Commit
5493527
1 Parent(s): 2f316e7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +10 -17
README.md CHANGED
@@ -13,12 +13,17 @@ model-index:
13
 
14
  This model is a fine-tuned version of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on a dataset of 30k newspaper headlines in German, Polish, English, Dutch and Spanish. The dataset contains 6k headlines in each of the five languages. The newspapers used are as follows:
15
 
16
- ['fakt', 'Rzeczpospolita', 'gazeta_wyborcza', 'UK_times',
17
- 'guardian', 'UK_sun', 'NRC', 'de_telegraaf', 'volkskrant',
18
- 'el_mundo', 'el_pais', 'ABC_spain', 'suddeutsche_zeitung',
19
- 'De_Welt', 'Bild']
 
20
 
 
21
 
 
 
 
22
 
23
 
24
  It achieves the following results on the evaluation set:
@@ -28,19 +33,6 @@ It achieves the following results on the evaluation set:
28
  - Validation Sparse Categorical Accuracy: 0.6434
29
  - Epoch: 4
30
 
31
- ## Model description
32
-
33
- More information needed
34
-
35
- ## Intended uses & limitations
36
-
37
- Newpaper headline classification
38
-
39
- ## Training and evaluation data
40
-
41
- More information needed
42
-
43
- ## Training procedure
44
 
45
  ### Training hyperparameters
46
 
@@ -64,3 +56,4 @@ The following hyperparameters were used during training:
64
  - Transformers 4.26.0
65
  - TensorFlow 2.9.2
66
  - Tokenizers 0.13.2
 
 
13
 
14
  This model is a fine-tuned version of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on a dataset of 30k newspaper headlines in German, Polish, English, Dutch and Spanish. The dataset contains 6k headlines in each of the five languages. The newspapers used are as follows:
15
 
16
+ Polish: Fakt, Rzeczpospolita, Gazeta Wyborcza
17
+ English: The Times, The Guardian, The Sun
18
+ Dutch: De Telegraaf, NRC, Volkskrant
19
+ Spanish: El Mundo, El Pais, ABC
20
+ German: Suddeutsche Zeitung, De Welt, Bild
21
 
22
+ coding scheme:
23
 
24
+ - "LABEL_0": negative,
25
+ - "LABEL_1": neutral,
26
+ - "LABEL_2": "positive"
27
 
28
 
29
  It achieves the following results on the evaluation set:
 
33
  - Validation Sparse Categorical Accuracy: 0.6434
34
  - Epoch: 4
35
 
 
 
 
 
 
 
 
 
 
 
 
 
 
36
 
37
  ### Training hyperparameters
38
 
 
56
  - Transformers 4.26.0
57
  - TensorFlow 2.9.2
58
  - Tokenizers 0.13.2
59
+