ophelielacroix commited on
Commit
d25b552
1 Parent(s): c561b98

First version of the da-bert-emotion-classification model and tokenizer

Browse files
README.md ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - da
4
+ tags:
5
+ - bert
6
+ - pytorch
7
+ - emotion
8
+ license: CC-BY_4.0
9
+ datasets:
10
+ - social media
11
+ metrics:
12
+ - f1
13
+ widget:
14
+ - text: "Jeg ejer en rød bil og det er en god bil."
15
+ ---
16
+
17
+ # Danish BERT for emotion classification
18
+
19
+ The BERT Emotion model classifies a Danish text in one of the following class:
20
+ * Glæde/Sindsro
21
+ * Tillid/Accept
22
+ * Forventning/Interrese
23
+ * Overasket/Målløs
24
+ * Vrede/Irritation
25
+ * Foragt/Modvilje
26
+ * Sorg/trist
27
+ * Frygt/Bekymret
28
+
29
+ It is based on the pretrained [Danish BERT](https://github.com/certainlyio/nordic_bert) model by BotXO which has been fine-tuned on social media data.
30
+
31
+ This model should be used after detecting whether the text contains emotion or not, using the binary [BERT Emotion model](https://huggingface.co/DaNLP/da-bert-emotion-binary).
32
+
33
+ See the [DaNLP documentation](https://danlp-alexandra.readthedocs.io/en/latest/docs/tasks/sentiment_analysis.html#bert-emotion) for more details.
34
+
35
+ Here is how to use the model:
36
+
37
+ ```python
38
+ from transformers import BertTokenizer, BertForSequenceClassification
39
+
40
+ model = BertForSequenceClassification.from_pretrained("DaNLP/da-bert-emotion-classification")
41
+ tokenizer = BertTokenizer.from_pretrained("DaNLP/da-bert-emotion-classification")
42
+ ```
43
+
44
+ ## Training data
45
+
46
+ The data used for training has not been made publicly available. It consists of social media data manually annotated in collaboration with Danmarks Radio.
47
+
added_tokens.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {}
config.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": ".",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "directionality": "bidi",
8
+ "finetuning_task": "emo",
9
+ "gradient_checkpointing": false,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "id2label": {
14
+ "0": "Gl\u00e6de/Sindsro",
15
+ "1": "Tillid/Accept",
16
+ "2": "Forventning/Interrese",
17
+ "3": "Overasket/M\u00e5ll\u00f8s",
18
+ "4": "Vrede/Irritation",
19
+ "5": "Foragt/Modvilje",
20
+ "6": "Sorg/trist",
21
+ "7": "Frygt/Bekymret"
22
+ },
23
+ "initializer_range": 0.02,
24
+ "intermediate_size": 3072,
25
+ "label2id": {
26
+ "Foragt/Modvilje": "5",
27
+ "Forventning/Interrese": "2",
28
+ "Frygt/Bekymret": "7",
29
+ "Gl\u00e6de/Sindsro": "0",
30
+ "Overasket/M\u00e5ll\u00f8s": "3",
31
+ "Sorg/trist": "6",
32
+ "Tillid/Accept": "1",
33
+ "Vrede/Irritation": "4"
34
+ },
35
+ "layer_norm_eps": 1e-12,
36
+ "max_position_embeddings": 512,
37
+ "model_type": "bert",
38
+ "num_attention_heads": 12,
39
+ "num_hidden_layers": 12,
40
+ "output_past": true,
41
+ "pad_token_id": 0,
42
+ "pooler_fc_size": 768,
43
+ "pooler_num_attention_heads": 12,
44
+ "pooler_num_fc_layers": 3,
45
+ "pooler_size_per_head": 128,
46
+ "pooler_type": "first_token_transform",
47
+ "type_vocab_size": 2,
48
+ "vocab_size": 32000
49
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1913b1dd404d8358e9b42444fb83339be802af550fa749a2cb82f73702941a8f
3
+ size 442580489
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:473a893298f80a1f7573d57d4df6ae0f78128c8205fb10ab3c85ad5792293619
3
+ size 442764672
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "init_inputs": []}
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8914ed596080bcc449b7e1ec71dc4cf483350fc108917ec2d757b826c72bdfcd
3
+ size 1257
vocab.txt ADDED
The diff for this file is too large to render. See raw diff