Merge branch 'main' of https://huggingface.co/wangfan/jdt-fin-roberta-wwm-large into main
Browse files- README.md +9 -0
- config.json +5 -6
README.md
CHANGED
@@ -1,3 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
在众多业务中,越来越频繁的使用预训练语言模型(Pre-trained Language Models),为了在金融场景下各任务中取得更好效果,我们发布了jdt-fin-roberta-wwm模型
|
2 |
|
3 |
## 模型
|
1 |
+
---
|
2 |
+
language: zh
|
3 |
+
tags:
|
4 |
+
- roberta-wwm
|
5 |
+
license: apache-2.0
|
6 |
+
datasets:
|
7 |
+
- finance
|
8 |
+
---
|
9 |
+
|
10 |
在众多业务中,越来越频繁的使用预训练语言模型(Pre-trained Language Models),为了在金融场景下各任务中取得更好效果,我们发布了jdt-fin-roberta-wwm模型
|
11 |
|
12 |
## 模型
|
config.json
CHANGED
@@ -3,8 +3,9 @@
|
|
3 |
"BertForMaskedLM"
|
4 |
],
|
5 |
"attention_probs_dropout_prob": 0.1,
|
|
|
6 |
"directionality": "bidi",
|
7 |
-
"
|
8 |
"hidden_act": "gelu",
|
9 |
"hidden_dropout_prob": 0.1,
|
10 |
"hidden_size": 1024,
|
@@ -15,15 +16,13 @@
|
|
15 |
"model_type": "bert",
|
16 |
"num_attention_heads": 16,
|
17 |
"num_hidden_layers": 24,
|
18 |
-
"
|
|
|
19 |
"pooler_fc_size": 768,
|
20 |
"pooler_num_attention_heads": 12,
|
21 |
"pooler_num_fc_layers": 3,
|
22 |
"pooler_size_per_head": 128,
|
23 |
"pooler_type": "first_token_transform",
|
24 |
-
"position_embedding_type": "absolute",
|
25 |
-
"transformers_version": "4.6.1",
|
26 |
"type_vocab_size": 2,
|
27 |
-
"use_cache": true,
|
28 |
"vocab_size": 21128
|
29 |
-
}
|
3 |
"BertForMaskedLM"
|
4 |
],
|
5 |
"attention_probs_dropout_prob": 0.1,
|
6 |
+
"bos_token_id": 0,
|
7 |
"directionality": "bidi",
|
8 |
+
"eos_token_id": 2,
|
9 |
"hidden_act": "gelu",
|
10 |
"hidden_dropout_prob": 0.1,
|
11 |
"hidden_size": 1024,
|
16 |
"model_type": "bert",
|
17 |
"num_attention_heads": 16,
|
18 |
"num_hidden_layers": 24,
|
19 |
+
"output_past": true,
|
20 |
+
"pad_token_id": 1,
|
21 |
"pooler_fc_size": 768,
|
22 |
"pooler_num_attention_heads": 12,
|
23 |
"pooler_num_fc_layers": 3,
|
24 |
"pooler_size_per_head": 128,
|
25 |
"pooler_type": "first_token_transform",
|
|
|
|
|
26 |
"type_vocab_size": 2,
|
|
|
27 |
"vocab_size": 21128
|
28 |
+
}
|