mk_core_news_md / README.md
1
---
2
tags:
3
- spacy
4
- token-classification
5
language:
6
- mk
7
license: CC-BY-SA-4.0
8
model-index:
9
- name: mk_core_news_md
10
  results:
11
  - task:
12
      name: NER
13
      type: token-classification
14
    metrics:
15
    - name: NER Precision
16
      type: precision
17
      value: 0.7577586207
18
    - name: NER Recall
19
      type: recall
20
      value: 0.7480851064
21
    - name: NER F Score
22
      type: f_score
23
      value: 0.7528907923
24
  - task:
25
      name: SENTER
26
      type: token-classification
27
    metrics:
28
    - name: SENTER Precision
29
      type: precision
30
      value: 0.768115942
31
    - name: SENTER Recall
32
      type: recall
33
      value: 0.6883116883
34
    - name: SENTER F Score
35
      type: f_score
36
      value: 0.7260273973
37
  - task:
38
      name: UNLABELED_DEPENDENCIES
39
      type: token-classification
40
    metrics:
41
    - name: Unlabeled Dependencies Accuracy
42
      type: accuracy
43
      value: 0.68633235
44
  - task:
45
      name: LABELED_DEPENDENCIES
46
      type: token-classification
47
    metrics:
48
    - name: Labeled Dependencies Accuracy
49
      type: accuracy
50
      value: 0.68633235
51
---
52
### Details: https://spacy.io/models/mk#mk_core_news_md
53
54
Macedonian pipeline optimized for CPU. Components: tok2vec, morphologizer, parser, senter, ner, attribute_ruler, lemmatizer.
55
56
| Feature | Description |
57
| --- | --- |
58
| **Name** | `mk_core_news_md` |
59
| **Version** | `3.1.0` |
60
| **spaCy** | `>=3.1.0,<3.2.0` |
61
| **Default Pipeline** | `morphologizer`, `parser`, `attribute_ruler`, `lemmatizer`, `ner` |
62
| **Components** | `morphologizer`, `parser`, `senter`, `attribute_ruler`, `lemmatizer`, `ner` |
63
| **Vectors** | 274587 keys, 20000 unique vectors (300 dimensions) |
64
| **Sources** | [Macedonian Corpus](https://blog.netcetera.com/macedonian-spacy-f3c85484777f) (Damjan Zlatinov, Melanija Gerasimovska, Borijan Georgievski, Marija Todosovska)<br />[Macedonian Corpus](https://blog.netcetera.com/macedonian-spacy-f3c85484777f) (Damjan Zlatinov, Melanija Gerasimovska, Borijan Georgievski, Marija Todosovska)<br />[Macedonian Corpus](https://blog.netcetera.com/macedonian-spacy-f3c85484777f) (Damjan Zlatinov, Melanija Gerasimovska, Borijan Georgievski, Marija Todosovska)<br />[spaCy lookups data](https://github.com/explosion/spacy-lookups-data) (Explosion)<br />[Explosion fastText Vectors (cbow, OSCAR Common Crawl + Wikipedia)](https://spacy.io) (Explosion) |
65
| **License** | `CC BY-SA 4.0` |
66
| **Author** | [Explosion](https://explosion.ai) |
67
68
### Label Scheme
69
70
<details>
71
72
<summary>View label scheme (55 labels for 4 components)</summary>
73
74
| Component | Labels |
75
| --- | --- |
76
| **`morphologizer`** | `POS=PROPN`, `POS=AUX`, `POS=ADJ`, `POS=NOUN`, `POS=ADP`, `POS=PUNCT`, `POS=CONJ`, `POS=NUM`, `POS=VERB`, `POS=PRON`, `POS=ADV`, `POS=SCONJ`, `POS=PART`, `POS=SYM`, `POS=X`, `_`, `POS=INTJ` |
77
| **`parser`** | `ROOT`, `advmod`, `att`, `aux`, `cc`, `dep`, `det`, `dobj`, `iobj`, `neg`, `nsubj`, `pobj`, `poss`, `pozm`, `pozv`, `prep`, `punct`, `relcl` |
78
| **`senter`** | `I`, `S` |
79
| **`ner`** | `CARDINAL`, `DATE`, `EVENT`, `FAC`, `GPE`, `LANGUAGE`, `LAW`, `LOC`, `MONEY`, `NORP`, `ORDINAL`, `ORG`, `PERCENT`, `PERSON`, `PRODUCT`, `QUANTITY`, `TIME`, `WORK_OF_ART` |
80
81
</details>
82
83
### Accuracy
84
85
| Type | Score |
86
| --- | --- |
87
| `TOKEN_ACC` | 100.00 |
88
| `POS_ACC` | 93.15 |
89
| `SENTS_P` | 76.81 |
90
| `SENTS_R` | 68.83 |
91
| `SENTS_F` | 72.60 |
92
| `DEP_UAS` | 68.63 |
93
| `DEP_LAS` | 53.29 |
94
| `ENTS_P` | 75.78 |
95
| `ENTS_R` | 74.81 |
96
| `ENTS_F` | 75.29 |
97