Update spaCy pipeline
Browse files- .gitattributes +1 -0
- README.md +14 -12
- __init__.py +10 -0
- meta.json +80 -64
- ner/model +2 -2
- ner/moves +1 -1
- pl_hacknationner-any-py3-none-any.whl +3 -0
- tok2vec/model +1 -1
- vocab/strings.json +0 -0
.gitattributes
CHANGED
|
@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 36 |
ner/model filter=lfs diff=lfs merge=lfs -text
|
| 37 |
pl_hacknationNER-any-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
|
| 38 |
tok2vec/model filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 36 |
ner/model filter=lfs diff=lfs merge=lfs -text
|
| 37 |
pl_hacknationNER-any-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
|
| 38 |
tok2vec/model filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
pl_hacknationner-any-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
|
@@ -6,7 +6,7 @@ language:
|
|
| 6 |
- pl
|
| 7 |
license: apache-2.0
|
| 8 |
model-index:
|
| 9 |
-
- name:
|
| 10 |
results:
|
| 11 |
- task:
|
| 12 |
name: NER
|
|
@@ -14,18 +14,20 @@ model-index:
|
|
| 14 |
metrics:
|
| 15 |
- name: NER Precision
|
| 16 |
type: precision
|
| 17 |
-
value: 0.
|
| 18 |
- name: NER Recall
|
| 19 |
type: recall
|
| 20 |
-
value: 0.
|
| 21 |
- name: NER F Score
|
| 22 |
type: f_score
|
| 23 |
-
value: 0.
|
| 24 |
---
|
|
|
|
|
|
|
| 25 |
| Feature | Description |
|
| 26 |
| --- | --- |
|
| 27 |
| **Name** | `pl_hacknationNER` |
|
| 28 |
-
| **Version** | `1.
|
| 29 |
| **spaCy** | `>=3.8.3,<3.9.0` |
|
| 30 |
| **Default Pipeline** | `tok2vec`, `ner` |
|
| 31 |
| **Components** | `tok2vec`, `ner` |
|
|
@@ -38,11 +40,11 @@ model-index:
|
|
| 38 |
|
| 39 |
<details>
|
| 40 |
|
| 41 |
-
<summary>View label scheme (
|
| 42 |
|
| 43 |
| Component | Labels |
|
| 44 |
| --- | --- |
|
| 45 |
-
| **`ner`** | `ADDRESS`, `AGE`, `BANK-ACCOUNT`, `CITY`, `COMPANY`, `DATE`, `DOCUMENT-NUMBER`, `EMAIL`, `HEALTH`, `JOB-TITLE`, `NAME`, `PESEL`, `PHONE`, `RELATIVE`, `SEX`, `SEXUAL-ORIENTATION`, `SURNAME` |
|
| 46 |
|
| 47 |
</details>
|
| 48 |
|
|
@@ -50,8 +52,8 @@ model-index:
|
|
| 50 |
|
| 51 |
| Type | Score |
|
| 52 |
| --- | --- |
|
| 53 |
-
| `ENTS_F` |
|
| 54 |
-
| `ENTS_P` |
|
| 55 |
-
| `ENTS_R` |
|
| 56 |
-
| `TOK2VEC_LOSS` |
|
| 57 |
-
| `NER_LOSS` |
|
|
|
|
| 6 |
- pl
|
| 7 |
license: apache-2.0
|
| 8 |
model-index:
|
| 9 |
+
- name: pl_hacknationner
|
| 10 |
results:
|
| 11 |
- task:
|
| 12 |
name: NER
|
|
|
|
| 14 |
metrics:
|
| 15 |
- name: NER Precision
|
| 16 |
type: precision
|
| 17 |
+
value: 0.7677725118
|
| 18 |
- name: NER Recall
|
| 19 |
type: recall
|
| 20 |
+
value: 0.6403162055
|
| 21 |
- name: NER F Score
|
| 22 |
type: f_score
|
| 23 |
+
value: 0.6982758621
|
| 24 |
---
|
| 25 |
+
second round of training
|
| 26 |
+
|
| 27 |
| Feature | Description |
|
| 28 |
| --- | --- |
|
| 29 |
| **Name** | `pl_hacknationNER` |
|
| 30 |
+
| **Version** | `1.2` |
|
| 31 |
| **spaCy** | `>=3.8.3,<3.9.0` |
|
| 32 |
| **Default Pipeline** | `tok2vec`, `ner` |
|
| 33 |
| **Components** | `tok2vec`, `ner` |
|
|
|
|
| 40 |
|
| 41 |
<details>
|
| 42 |
|
| 43 |
+
<summary>View label scheme (23 labels for 1 components)</summary>
|
| 44 |
|
| 45 |
| Component | Labels |
|
| 46 |
| --- | --- |
|
| 47 |
+
| **`ner`** | `ADDRESS`, `AGE`, `BANK-ACCOUNT`, `CITY`, `COMPANY`, `DATE`, `DOCUMENT-NUMBER`, `EMAIL`, `ETHNICITY`, `HEALTH`, `JOB-TITLE`, `NAME`, `PESEL`, `PHONE`, `POLITICAL-VIEW`, `RELATIVE`, `RELIGION`, `SCHOOL-NAME`, `SECRET`, `SEX`, `SEXUAL-ORIENTATION`, `SURNAME`, `USERNAME` |
|
| 48 |
|
| 49 |
</details>
|
| 50 |
|
|
|
|
| 52 |
|
| 53 |
| Type | Score |
|
| 54 |
| --- | --- |
|
| 55 |
+
| `ENTS_F` | 69.83 |
|
| 56 |
+
| `ENTS_P` | 76.78 |
|
| 57 |
+
| `ENTS_R` | 64.03 |
|
| 58 |
+
| `TOK2VEC_LOSS` | 58885.11 |
|
| 59 |
+
| `NER_LOSS` | 17819.95 |
|
__init__.py
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from pathlib import Path
|
| 2 |
+
from spacy.util import load_model_from_init_py, get_model_meta
|
| 3 |
+
|
| 4 |
+
|
| 5 |
+
|
| 6 |
+
__version__ = get_model_meta(Path(__file__).parent)['version']
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
def load(**overrides):
|
| 10 |
+
return load_model_from_init_py(__file__, **overrides)
|
meta.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"lang":"pl",
|
| 3 |
"name":"hacknationNER",
|
| 4 |
-
"version":"1.
|
| 5 |
-
"description":"",
|
| 6 |
"author":"Ariel",
|
| 7 |
"email":"ariel.m.drozd@gmail.com",
|
| 8 |
"url":"",
|
|
@@ -28,15 +28,21 @@
|
|
| 28 |
"DATE",
|
| 29 |
"DOCUMENT-NUMBER",
|
| 30 |
"EMAIL",
|
|
|
|
| 31 |
"HEALTH",
|
| 32 |
"JOB-TITLE",
|
| 33 |
"NAME",
|
| 34 |
"PESEL",
|
| 35 |
"PHONE",
|
|
|
|
| 36 |
"RELATIVE",
|
|
|
|
|
|
|
|
|
|
| 37 |
"SEX",
|
| 38 |
"SEXUAL-ORIENTATION",
|
| 39 |
-
"SURNAME"
|
|
|
|
| 40 |
]
|
| 41 |
},
|
| 42 |
"pipeline":[
|
|
@@ -51,98 +57,108 @@
|
|
| 51 |
|
| 52 |
],
|
| 53 |
"performance":{
|
| 54 |
-
"ents_f":0.
|
| 55 |
-
"ents_p":0.
|
| 56 |
-
"ents_r":0.
|
| 57 |
"ents_per_type":{
|
| 58 |
-
"
|
| 59 |
-
"p":0.
|
| 60 |
-
"r":0.
|
| 61 |
-
"f":0.
|
| 62 |
},
|
| 63 |
-
"
|
| 64 |
-
"p":0.
|
| 65 |
-
"r":0.
|
| 66 |
-
"f":0.
|
| 67 |
},
|
| 68 |
-
"
|
| 69 |
-
"p":0.
|
| 70 |
-
"r":0.
|
| 71 |
-
"f":0.
|
| 72 |
},
|
| 73 |
-
"
|
| 74 |
-
"p":
|
| 75 |
-
"r":0.
|
| 76 |
-
"f":0.
|
| 77 |
},
|
| 78 |
-
"
|
| 79 |
-
"p":0.
|
| 80 |
-
"r":0.
|
| 81 |
-
"f":0.
|
| 82 |
},
|
| 83 |
-
"
|
| 84 |
-
"p":0.
|
| 85 |
-
"r":0.
|
| 86 |
-
"f":0.
|
| 87 |
},
|
| 88 |
-
"
|
| 89 |
-
"p":0.
|
| 90 |
-
"r":0.
|
| 91 |
-
"f":0.
|
| 92 |
},
|
| 93 |
-
"
|
| 94 |
-
"p":
|
| 95 |
-
"r":
|
| 96 |
-
"f":
|
| 97 |
},
|
| 98 |
"AGE":{
|
| 99 |
-
"p":0.
|
| 100 |
-
"r":
|
| 101 |
-
"f":0.
|
| 102 |
},
|
| 103 |
-
"
|
| 104 |
"p":1.0,
|
| 105 |
-
"r":
|
| 106 |
-
"f":
|
| 107 |
},
|
| 108 |
-
"
|
| 109 |
-
"p":0.
|
| 110 |
-
"r":0.
|
| 111 |
-
"f":0.
|
| 112 |
},
|
| 113 |
-
"
|
| 114 |
-
"p":
|
| 115 |
-
"r":0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 116 |
"f":0.5
|
| 117 |
},
|
| 118 |
-
"
|
| 119 |
-
"p":0.
|
| 120 |
-
"r":0.
|
| 121 |
-
"f":0.
|
| 122 |
},
|
| 123 |
-
"
|
| 124 |
"p":0.0,
|
| 125 |
"r":0.0,
|
| 126 |
"f":0.0
|
| 127 |
},
|
| 128 |
-
"
|
| 129 |
-
"p":
|
| 130 |
-
"r":0.
|
| 131 |
-
"f":0.
|
| 132 |
},
|
| 133 |
-
"
|
| 134 |
"p":0.0,
|
| 135 |
"r":0.0,
|
| 136 |
"f":0.0
|
| 137 |
},
|
| 138 |
-
"
|
| 139 |
"p":1.0,
|
| 140 |
"r":0.3333333333,
|
| 141 |
"f":0.5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 142 |
}
|
| 143 |
},
|
| 144 |
-
"tok2vec_loss":
|
| 145 |
-
"ner_loss":
|
| 146 |
},
|
| 147 |
"requirements":[
|
| 148 |
"spacy>=3.8.3,<3.9.0"
|
|
|
|
| 1 |
{
|
| 2 |
"lang":"pl",
|
| 3 |
"name":"hacknationNER",
|
| 4 |
+
"version":"1.2",
|
| 5 |
+
"description":"second round of training",
|
| 6 |
"author":"Ariel",
|
| 7 |
"email":"ariel.m.drozd@gmail.com",
|
| 8 |
"url":"",
|
|
|
|
| 28 |
"DATE",
|
| 29 |
"DOCUMENT-NUMBER",
|
| 30 |
"EMAIL",
|
| 31 |
+
"ETHNICITY",
|
| 32 |
"HEALTH",
|
| 33 |
"JOB-TITLE",
|
| 34 |
"NAME",
|
| 35 |
"PESEL",
|
| 36 |
"PHONE",
|
| 37 |
+
"POLITICAL-VIEW",
|
| 38 |
"RELATIVE",
|
| 39 |
+
"RELIGION",
|
| 40 |
+
"SCHOOL-NAME",
|
| 41 |
+
"SECRET",
|
| 42 |
"SEX",
|
| 43 |
"SEXUAL-ORIENTATION",
|
| 44 |
+
"SURNAME",
|
| 45 |
+
"USERNAME"
|
| 46 |
]
|
| 47 |
},
|
| 48 |
"pipeline":[
|
|
|
|
| 57 |
|
| 58 |
],
|
| 59 |
"performance":{
|
| 60 |
+
"ents_f":0.6982758621,
|
| 61 |
+
"ents_p":0.7677725118,
|
| 62 |
+
"ents_r":0.6403162055,
|
| 63 |
"ents_per_type":{
|
| 64 |
+
"DATE":{
|
| 65 |
+
"p":0.875,
|
| 66 |
+
"r":0.7,
|
| 67 |
+
"f":0.7777777778
|
| 68 |
},
|
| 69 |
+
"PESEL":{
|
| 70 |
+
"p":0.7777777778,
|
| 71 |
+
"r":0.6363636364,
|
| 72 |
+
"f":0.7
|
| 73 |
},
|
| 74 |
+
"DOCUMENT-NUMBER":{
|
| 75 |
+
"p":0.8333333333,
|
| 76 |
+
"r":0.652173913,
|
| 77 |
+
"f":0.7317073171
|
| 78 |
},
|
| 79 |
+
"JOB-TITLE":{
|
| 80 |
+
"p":0.625,
|
| 81 |
+
"r":0.2777777778,
|
| 82 |
+
"f":0.3846153846
|
| 83 |
},
|
| 84 |
+
"EMAIL":{
|
| 85 |
+
"p":0.8333333333,
|
| 86 |
+
"r":0.9375,
|
| 87 |
+
"f":0.8823529412
|
| 88 |
},
|
| 89 |
+
"PHONE":{
|
| 90 |
+
"p":0.6818181818,
|
| 91 |
+
"r":0.8823529412,
|
| 92 |
+
"f":0.7692307692
|
| 93 |
},
|
| 94 |
+
"NAME":{
|
| 95 |
+
"p":0.7307692308,
|
| 96 |
+
"r":0.76,
|
| 97 |
+
"f":0.7450980392
|
| 98 |
},
|
| 99 |
+
"CITY":{
|
| 100 |
+
"p":0.7142857143,
|
| 101 |
+
"r":0.6666666667,
|
| 102 |
+
"f":0.6896551724
|
| 103 |
},
|
| 104 |
"AGE":{
|
| 105 |
+
"p":0.8125,
|
| 106 |
+
"r":0.8666666667,
|
| 107 |
+
"f":0.8387096774
|
| 108 |
},
|
| 109 |
+
"SEX":{
|
| 110 |
"p":1.0,
|
| 111 |
+
"r":0.8181818182,
|
| 112 |
+
"f":0.9
|
| 113 |
},
|
| 114 |
+
"ADDRESS":{
|
| 115 |
+
"p":0.7368421053,
|
| 116 |
+
"r":0.6086956522,
|
| 117 |
+
"f":0.6666666667
|
| 118 |
},
|
| 119 |
+
"SURNAME":{
|
| 120 |
+
"p":0.8888888889,
|
| 121 |
+
"r":0.6956521739,
|
| 122 |
+
"f":0.7804878049
|
| 123 |
+
},
|
| 124 |
+
"RELATIVE":{
|
| 125 |
+
"p":0.8,
|
| 126 |
+
"r":0.3636363636,
|
| 127 |
"f":0.5
|
| 128 |
},
|
| 129 |
+
"COMPANY":{
|
| 130 |
+
"p":0.5,
|
| 131 |
+
"r":0.1428571429,
|
| 132 |
+
"f":0.2222222222
|
| 133 |
},
|
| 134 |
+
"HEALTH":{
|
| 135 |
"p":0.0,
|
| 136 |
"r":0.0,
|
| 137 |
"f":0.0
|
| 138 |
},
|
| 139 |
+
"POLITICAL-VIEW":{
|
| 140 |
+
"p":0.5,
|
| 141 |
+
"r":0.25,
|
| 142 |
+
"f":0.3333333333
|
| 143 |
},
|
| 144 |
+
"SCHOOL-NAME":{
|
| 145 |
"p":0.0,
|
| 146 |
"r":0.0,
|
| 147 |
"f":0.0
|
| 148 |
},
|
| 149 |
+
"ETHNICITY":{
|
| 150 |
"p":1.0,
|
| 151 |
"r":0.3333333333,
|
| 152 |
"f":0.5
|
| 153 |
+
},
|
| 154 |
+
"BANK-ACCOUNT":{
|
| 155 |
+
"p":0.0,
|
| 156 |
+
"r":0.0,
|
| 157 |
+
"f":0.0
|
| 158 |
}
|
| 159 |
},
|
| 160 |
+
"tok2vec_loss":588.851107855,
|
| 161 |
+
"ner_loss":178.1994781494
|
| 162 |
},
|
| 163 |
"requirements":[
|
| 164 |
"spacy>=3.8.3,<3.9.0"
|
ner/model
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87cd3fafd03834601896c8606cf86bbf1dc8336ac35a9e175d69ff4633bb33e8
|
| 3 |
+
size 150389
|
ner/moves
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
��moves�
|
|
|
|
| 1 |
+
��moves�0{"0":{},"1":{"ADDRESS":625,"PHONE":317,"DOCUMENT-NUMBER":173,"CITY":153,"NAME":146,"COMPANY":135,"SURNAME":106,"EMAIL":103,"AGE":90,"JOB-TITLE":72,"PESEL":65,"DATE":59,"SCHOOL-NAME":58,"SEX":49,"HEALTH":41,"RELATIVE":37,"USERNAME":21,"POLITICAL-VIEW":13,"SEXUAL-ORIENTATION":11,"ETHNICITY":7,"RELIGION":5,"BANK-ACCOUNT":5,"SECRET":2},"2":{"ADDRESS":625,"PHONE":317,"DOCUMENT-NUMBER":173,"CITY":153,"NAME":146,"COMPANY":135,"SURNAME":106,"EMAIL":103,"AGE":90,"JOB-TITLE":72,"PESEL":65,"DATE":59,"SCHOOL-NAME":58,"SEX":49,"HEALTH":41,"RELATIVE":37,"USERNAME":21,"POLITICAL-VIEW":13,"SEXUAL-ORIENTATION":11,"ETHNICITY":7,"RELIGION":5,"BANK-ACCOUNT":5,"SECRET":2},"3":{"ADDRESS":625,"PHONE":317,"DOCUMENT-NUMBER":173,"CITY":153,"NAME":146,"COMPANY":135,"SURNAME":106,"EMAIL":103,"AGE":90,"JOB-TITLE":72,"PESEL":65,"DATE":59,"SCHOOL-NAME":58,"SEX":49,"HEALTH":41,"RELATIVE":37,"USERNAME":21,"POLITICAL-VIEW":13,"SEXUAL-ORIENTATION":11,"ETHNICITY":7,"RELIGION":5,"BANK-ACCOUNT":5,"SECRET":2},"4":{"ADDRESS":625,"PHONE":317,"DOCUMENT-NUMBER":173,"CITY":153,"NAME":146,"COMPANY":135,"SURNAME":106,"EMAIL":103,"AGE":90,"JOB-TITLE":72,"PESEL":65,"DATE":59,"SCHOOL-NAME":58,"SEX":49,"HEALTH":41,"RELATIVE":37,"USERNAME":21,"POLITICAL-VIEW":13,"SEXUAL-ORIENTATION":11,"ETHNICITY":7,"RELIGION":5,"BANK-ACCOUNT":5,"SECRET":2,"":1},"5":{"":1}}�cfg��neg_key�
|
pl_hacknationner-any-py3-none-any.whl
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a1f579d158191ffc82ea68254ba8984e392a09e04447220bd45f13769b4a7726
|
| 3 |
+
size 5725355
|
tok2vec/model
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6009091
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c805a95cdf348d2cd142c09a0adbcbefaceb42e32161f8ed8e408372298a6ba3
|
| 3 |
size 6009091
|
vocab/strings.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|