Update README.md
Browse files
README.md
CHANGED
@@ -11,11 +11,9 @@ size_categories:
|
|
11 |
|
12 |
## Dataset Description
|
13 |
|
14 |
-
- **Homepage:**
|
15 |
-
- **
|
16 |
-
- **
|
17 |
-
- **Leaderboard:**
|
18 |
-
- **Point of Contact:**
|
19 |
|
20 |
### Dataset Summary
|
21 |
|
@@ -23,84 +21,30 @@ Este Dataset foi criado à partir dos dados disponibilizados pelo Grupo de Proce
|
|
23 |
|
24 |
### Supported Tasks and Leaderboards
|
25 |
|
26 |
-
|
27 |
|
28 |
### Languages
|
29 |
|
30 |
-
|
31 |
-
|
32 |
-
## Dataset Structure
|
33 |
-
|
34 |
-
### Data Instances
|
35 |
-
|
36 |
-
[More Information Needed]
|
37 |
-
|
38 |
-
### Data Fields
|
39 |
-
|
40 |
-
[More Information Needed]
|
41 |
-
|
42 |
-
### Data Splits
|
43 |
-
|
44 |
-
[More Information Needed]
|
45 |
-
|
46 |
-
## Dataset Creation
|
47 |
-
|
48 |
-
### Curation Rationale
|
49 |
-
|
50 |
-
[More Information Needed]
|
51 |
-
|
52 |
-
### Source Data
|
53 |
|
54 |
#### Initial Data Collection and Normalization
|
55 |
|
56 |
-
|
57 |
-
|
58 |
-
#### Who are the source language producers?
|
59 |
-
|
60 |
-
[More Information Needed]
|
61 |
-
|
62 |
-
### Annotations
|
63 |
-
|
64 |
-
#### Annotation process
|
65 |
-
|
66 |
-
[More Information Needed]
|
67 |
|
68 |
-
|
69 |
-
|
70 |
-
[More Information Needed]
|
71 |
-
|
72 |
-
### Personal and Sensitive Information
|
73 |
-
|
74 |
-
[More Information Needed]
|
75 |
-
|
76 |
-
## Considerations for Using the Data
|
77 |
-
|
78 |
-
### Social Impact of Dataset
|
79 |
-
|
80 |
-
[More Information Needed]
|
81 |
-
|
82 |
-
### Discussion of Biases
|
83 |
-
|
84 |
-
[More Information Needed]
|
85 |
-
|
86 |
-
### Other Known Limitations
|
87 |
-
|
88 |
-
[More Information Needed]
|
89 |
-
|
90 |
-
## Additional Information
|
91 |
-
|
92 |
-
### Dataset Curators
|
93 |
|
94 |
-
|
95 |
|
96 |
-
###
|
97 |
|
98 |
-
|
99 |
|
100 |
-
###
|
101 |
|
102 |
-
|
103 |
|
104 |
-
|
|
|
|
|
105 |
|
106 |
-
|
|
|
11 |
|
12 |
## Dataset Description
|
13 |
|
14 |
+
- **Homepage:** https://www.inf.pucrs.br/linatural/wordpress/recursos-e-ferramentas/blogset-br/
|
15 |
+
- **Leaderboard:** Grupo de Processamento da Linguagem Natural da PUC-RS
|
16 |
+
- **Point of Contact:** Site oficial
|
|
|
|
|
17 |
|
18 |
### Dataset Summary
|
19 |
|
|
|
21 |
|
22 |
### Supported Tasks and Leaderboards
|
23 |
|
24 |
+
Indicado para treinamento de modelos de linguagem.
|
25 |
|
26 |
### Languages
|
27 |
|
28 |
+
Português do Brasil
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
29 |
|
30 |
#### Initial Data Collection and Normalization
|
31 |
|
32 |
+
Informações sobre a criação do Dataset podem ser encontradas aqui: https://www.inf.pucrs.br/linatural/wordpress/recursos-e-ferramentas/blogset-br/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
|
34 |
+
### Licensing Information
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
|
36 |
+
Apache V2
|
37 |
|
38 |
+
### Contributions
|
39 |
|
40 |
+
Esta página é meramente uma configuração para o formato Huggingface do trabalho realizado pelo equipe PLN da PUC-RS.
|
41 |
|
42 |
+
### Huggingface format
|
43 |
|
44 |
+
O código a seguir foi utilizado para a criação do dataset. Decisões quanto a estrutura:
|
45 |
|
46 |
+
1. Somente a coluna relacionada ao texto foi utilizada (coluna 4).
|
47 |
+
2. Foi aplicada uma bateria de ajustes visando limpar o texto conforme pode ser observado no código.
|
48 |
+
3. Procurou-se manter o limite de 512 palavras em cada linha.
|
49 |
|
50 |
+
Gist: https://gist.github.com/rdemorais/ce2e708af4c07aba47930bc12ed92472
|