Yehor Smoliakov commited on
Commit
61003b9
1 Parent(s): cd04ded
Files changed (6) hide show
  1. .gitignore +1 -0
  2. README.md +39 -3
  3. config.json +108 -0
  4. model.safetensors +3 -0
  5. tokenizer_config.json +1 -0
  6. vocab.txt +0 -0
.gitignore ADDED
@@ -0,0 +1 @@
 
 
1
+ .DS_Store
README.md CHANGED
@@ -1,3 +1,39 @@
1
- ---
2
- license: cc-by-nc-sa-4.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: uk
3
+
4
+ datasets: Yehor/ual-topics
5
+
6
+ license: cc-by-nc-sa-4.0
7
+ ---
8
+
9
+ ## Metrics
10
+
11
+ - Epochs: 50
12
+ - Test accuracy: 0.62500
13
+ - F1 (MACRO): 0.46573
14
+
15
+ ## How to use
16
+
17
+ ```python
18
+ import torch
19
+ from transformers import AutoTokenizer, AutoModel
20
+
21
+ tokenizer = AutoTokenizer.from_pretrained("Yehor/ual-topics-classifier")
22
+ model = AutoModel.from_pretrained("Yehor/ual-topics-classifier")
23
+
24
+ device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
25
+
26
+ topic_classifier = pipeline(task='text-classification', model=model, tokenizer=tokenizer, device=device)
27
+
28
+ question = """
29
+ Що мені робити на ВЛК
30
+ """
31
+
32
+ print(topic_classifier(question))
33
+
34
+ question = """
35
+ Які мої дії для отримання аліментів на дитину
36
+ """
37
+
38
+ print(topic_classifier(question))
39
+ ```
config.json ADDED
@@ -0,0 +1,108 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Geotrend/distilbert-base-uk-cased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "inshe",
13
+ "1": "ekologiya",
14
+ "2": "ziemielnie_pravo",
15
+ "3": "reklama",
16
+ "4": "bankivska_diialnist",
17
+ "5": "prava_spozhivachiv",
18
+ "6": "medicina",
19
+ "7": "spadkove_pravo",
20
+ "8": "immighratsiia_iemighratsiia",
21
+ "9": "intieliektualna_vlasnist",
22
+ "10": "gospodarskie_pravo",
23
+ "11": "pidpriemnicka_dialnist",
24
+ "12": "opodatkuvannia",
25
+ "13": "piensiiata_sotsialni_viplati",
26
+ "14": "viiskovie_pravo",
27
+ "15": "sudova_praktika",
28
+ "16": "kriminalnie_pravo",
29
+ "17": "gromadianski_pravovidnosini",
30
+ "18": "strakhuvannya",
31
+ "19": "pratsevlashtuvvannya",
32
+ "20": "sotsialnyj_zakhist",
33
+ "21": "vighotovliennia_produktsiyi_ta_nadannia_poslugh",
34
+ "22": "litsienzuvannia",
35
+ "23": "reyestraciya_likvidaciya_bankrutstvo",
36
+ "24": "doghovirni_vidnosini",
37
+ "25": "administrativnie_pravo",
38
+ "26": "nierukhomist",
39
+ "27": "prava_vnutrishno_pieriemishchienikh_osib",
40
+ "28": "investitsii",
41
+ "29": "notarialni_pytanniia",
42
+ "30": "avtovlasnykam",
43
+ "31": "zhitlovi_pravovidnosini",
44
+ "32": "dovircha_vlastnist",
45
+ "33": "dierzhavni_zakupivli",
46
+ "34": "simejne_pravo",
47
+ "35": "mytne_pravo",
48
+ "36": "mizhnarodni_pravovidnosini",
49
+ "37": "korporativnie_pravo",
50
+ "38": "tsivilne_pravo"
51
+ },
52
+ "initializer_range": 0.02,
53
+ "label2id": {
54
+ "administrativnie_pravo": 25,
55
+ "avtovlasnykam": 30,
56
+ "bankivska_diialnist": 4,
57
+ "dierzhavni_zakupivli": 33,
58
+ "doghovirni_vidnosini": 24,
59
+ "dovircha_vlastnist": 32,
60
+ "ekologiya": 1,
61
+ "gospodarskie_pravo": 10,
62
+ "gromadianski_pravovidnosini": 17,
63
+ "immighratsiia_iemighratsiia": 8,
64
+ "inshe": 0,
65
+ "intieliektualna_vlasnist": 9,
66
+ "investitsii": 28,
67
+ "korporativnie_pravo": 37,
68
+ "kriminalnie_pravo": 16,
69
+ "litsienzuvannia": 22,
70
+ "medicina": 6,
71
+ "mizhnarodni_pravovidnosini": 36,
72
+ "mytne_pravo": 35,
73
+ "nierukhomist": 26,
74
+ "notarialni_pytanniia": 29,
75
+ "opodatkuvannia": 12,
76
+ "pidpriemnicka_dialnist": 11,
77
+ "piensiiata_sotsialni_viplati": 13,
78
+ "pratsevlashtuvvannya": 19,
79
+ "prava_spozhivachiv": 5,
80
+ "prava_vnutrishno_pieriemishchienikh_osib": 27,
81
+ "reklama": 3,
82
+ "reyestraciya_likvidaciya_bankrutstvo": 23,
83
+ "simejne_pravo": 34,
84
+ "sotsialnyj_zakhist": 20,
85
+ "spadkove_pravo": 7,
86
+ "strakhuvannya": 18,
87
+ "sudova_praktika": 15,
88
+ "tsivilne_pravo": 38,
89
+ "vighotovliennia_produktsiyi_ta_nadannia_poslugh": 21,
90
+ "viiskovie_pravo": 14,
91
+ "zhitlovi_pravovidnosini": 31,
92
+ "ziemielnie_pravo": 2
93
+ },
94
+ "max_position_embeddings": 512,
95
+ "model_type": "distilbert",
96
+ "n_heads": 12,
97
+ "n_layers": 6,
98
+ "output_past": true,
99
+ "pad_token_id": 0,
100
+ "problem_type": "single_label_classification",
101
+ "qa_dropout": 0.1,
102
+ "seq_classif_dropout": 0.2,
103
+ "sinusoidal_pos_embds": false,
104
+ "tie_weights_": true,
105
+ "torch_dtype": "float32",
106
+ "transformers_version": "4.42.4",
107
+ "vocab_size": 11811
108
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44424169e0a108925e133c67817b3053b2db9de8e14400f0279cb1663ee02d69
3
+ size 210466132
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": false, "model_max_length": 512}
vocab.txt ADDED
The diff for this file is too large to render. See raw diff