silviapasuarez commited on
Commit
eaa054e
·
verified ·
1 Parent(s): fe04ecb

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +102 -0
README.md ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - gl
4
+ pretty_name: OASST2 Galician Subset
5
+ task_categories:
6
+ - text-generation
7
+ task_ids:
8
+ - dialogue-generation
9
+ tags:
10
+ - galician
11
+ - instruction-tuning
12
+ - chat
13
+ - conversation
14
+ - openassistant
15
+ - translation
16
+ license: apache-2.0
17
+ size_categories:
18
+ - 1K<n<10K
19
+ ---
20
+
21
+ # OASST2 Galician Subset
22
+
23
+ ## Dataset description
24
+
25
+ This dataset is a Galician translation/adaptation of a subset of the [OASST2](OpenAssistant/oasst2) conversational dataset. It is intended for instruction tuning, dialogue modeling, and related experiments in Galician.
26
+
27
+ This release contains 1,786 instances in JSONL format. It does not include the full original OASST2 dataset. The data preserves the original conversation-oriented structure, where messages are linked through tree and parent identifiers.
28
+
29
+ ## Dataset structure
30
+
31
+ The dataset is distributed in JSONL format. Each line contains one message node with the following fields:
32
+
33
+ - `message_tree_id`: identifier of the conversation tree
34
+ - `message_id`: identifier of the current message
35
+ - `parent_id`: identifier of the parent message; empty for root messages
36
+ - `lang`: language code
37
+ - `role`: speaker role, typically `prompter` or `assistant`
38
+ - `text`: message content in Galician
39
+
40
+ ### Example
41
+
42
+ ```json
43
+ {
44
+ "message_tree_id": "c55f670b-f384-48b0-ba71-e5a2b2c9137e",
45
+ "message_id": "c55f670b-f384-48b0-ba71-e5a2b2c9137e",
46
+ "parent_id": "",
47
+ "lang": "gl",
48
+ "role": "prompter",
49
+ "text": "Crea un bloque de estatísticas para un poderoso monstro de tipo morto vivente en Dragóns e Alxubes quinta edición."
50
+ }
51
+ ```
52
+
53
+ ## Data source and creation
54
+
55
+ This dataset is based on a subset of the original [OASST2](OpenAssistant/oasst2) dataset and was translated/adapted into Galician. It preserves the message-level conversational structure of the source data, including tree-level and parent-child relationships between turns.
56
+
57
+ The main purpose of this version is to provide conversational and instruction-following data in Galician for experimentation, fine-tuning, and evaluation in low-resource settings.
58
+
59
+ ## Intended uses
60
+
61
+ This dataset can be used for:
62
+
63
+ - conversational fine-tuning in Galician
64
+ - dialogue generation
65
+ - instruction tuning for chat-oriented models
66
+ - multilingual or cross-lingual experiments
67
+ - low-resource NLP research
68
+
69
+ ## Limitations
70
+
71
+ - This dataset is only a subset of the original OASST2 data.
72
+ - Since this is a translated/adapted version, some examples may reflect translation choices, stylistic variation, or localized phrasing relative to the source dataset.
73
+ - The dataset is structured at the message level, so conversation trees may need to be reconstructed programmatically for some use cases.
74
+ - The quality of the data depends on the translation/adaptation process used.
75
+
76
+ ## Licensing
77
+
78
+ This dataset follows the same license as the original OASST2 dataset: Apache License 2.0.
79
+
80
+ ## Usage
81
+
82
+ Example with `datasets`:
83
+
84
+ ```python
85
+ from datasets import load_dataset
86
+
87
+ ds = load_dataset("json", data_files="oasst2_gl_subset.jsonl")
88
+ print(ds["train"][0])
89
+ ```
90
+
91
+ If you want to reconstruct conversations by tree:
92
+
93
+ ```python
94
+ from datasets import load_dataset
95
+
96
+ ds = load_dataset("json", data_files="oasst2_gl_subset.jsonl")["train"]
97
+ print(ds[0]["message_tree_id"], ds[0]["role"], ds[0]["text"])
98
+ ```
99
+
100
+ ## Acknowledgements
101
+
102
+ This work is funded by the Ministerio para la Transformación Digital y de la Función Pública - Funded by EU – NextGenerationEU within the framework of the project Desarrollo de Modelos ALIA. (Esta publicación del proyecto Desarrollo de Modelos ALIA está financiada por el Ministerio para la Transformación Digital y de la Función Pública y por el Plan de Recuperación, Transformación y Resiliencia – Financiado por la Unión Europea – NextGenerationEU)