adrianeboyd
commited on
Commit
•
0f87523
1
Parent(s):
6cd4e04
Update spaCy pipeline
Browse files- README.md +2 -2
- accuracy.json +1 -1
- config.cfg +2 -0
- meta.json +4 -4
- morphologizer/cfg +1 -0
- pl_core_news_md-any-py3-none-any.whl +2 -2
- tagger/cfg +1 -0
README.md
CHANGED
@@ -78,8 +78,8 @@ Polish pipeline optimized for CPU. Components: tok2vec, morphologizer, parser, l
|
|
78 |
| Feature | Description |
|
79 |
| --- | --- |
|
80 |
| **Name** | `pl_core_news_md` |
|
81 |
-
| **Version** | `3.
|
82 |
-
| **spaCy** | `>=3.
|
83 |
| **Default Pipeline** | `tok2vec`, `morphologizer`, `parser`, `lemmatizer`, `tagger`, `attribute_ruler`, `ner` |
|
84 |
| **Components** | `tok2vec`, `morphologizer`, `parser`, `lemmatizer`, `tagger`, `senter`, `attribute_ruler`, `ner` |
|
85 |
| **Vectors** | 500000 keys, 20000 unique vectors (300 dimensions) |
|
|
|
78 |
| Feature | Description |
|
79 |
| --- | --- |
|
80 |
| **Name** | `pl_core_news_md` |
|
81 |
+
| **Version** | `3.6.0` |
|
82 |
+
| **spaCy** | `>=3.6.0,<3.7.0` |
|
83 |
| **Default Pipeline** | `tok2vec`, `morphologizer`, `parser`, `lemmatizer`, `tagger`, `attribute_ruler`, `ner` |
|
84 |
| **Components** | `tok2vec`, `morphologizer`, `parser`, `lemmatizer`, `tagger`, `senter`, `attribute_ruler`, `ner` |
|
85 |
| **Vectors** | 500000 keys, 20000 unique vectors (300 dimensions) |
|
accuracy.json
CHANGED
@@ -524,5 +524,5 @@
|
|
524 |
"f": 0.6502463054
|
525 |
}
|
526 |
},
|
527 |
-
"speed":
|
528 |
}
|
|
|
524 |
"f": 0.6502463054
|
525 |
}
|
526 |
},
|
527 |
+
"speed": 7821.6133862982
|
528 |
}
|
config.cfg
CHANGED
@@ -46,6 +46,7 @@ upstream = "tok2vec"
|
|
46 |
[components.morphologizer]
|
47 |
factory = "morphologizer"
|
48 |
extend = false
|
|
|
49 |
overwrite = true
|
50 |
scorer = {"@scorers":"spacy.morphologizer_scorer.v1"}
|
51 |
|
@@ -143,6 +144,7 @@ maxout_pieces = 2
|
|
143 |
|
144 |
[components.tagger]
|
145 |
factory = "tagger"
|
|
|
146 |
neg_prefix = "!"
|
147 |
overwrite = false
|
148 |
scorer = {"@scorers":"spacy.tagger_scorer.v1"}
|
|
|
46 |
[components.morphologizer]
|
47 |
factory = "morphologizer"
|
48 |
extend = false
|
49 |
+
label_smoothing = 0.0
|
50 |
overwrite = true
|
51 |
scorer = {"@scorers":"spacy.morphologizer_scorer.v1"}
|
52 |
|
|
|
144 |
|
145 |
[components.tagger]
|
146 |
factory = "tagger"
|
147 |
+
label_smoothing = 0.0
|
148 |
neg_prefix = "!"
|
149 |
overwrite = false
|
150 |
scorer = {"@scorers":"spacy.tagger_scorer.v1"}
|
meta.json
CHANGED
@@ -1,14 +1,14 @@
|
|
1 |
{
|
2 |
"lang":"pl",
|
3 |
"name":"core_news_md",
|
4 |
-
"version":"3.
|
5 |
"description":"Polish pipeline optimized for CPU. Components: tok2vec, morphologizer, parser, lemmatizer (trainable_lemmatizer), tagger, senter, ner.",
|
6 |
"author":"Explosion",
|
7 |
"email":"contact@explosion.ai",
|
8 |
"url":"https://explosion.ai",
|
9 |
"license":"GNU GPL 3.0",
|
10 |
-
"spacy_version":">=3.
|
11 |
-
"spacy_git_version":"
|
12 |
"vectors":{
|
13 |
"width":300,
|
14 |
"vectors":20000,
|
@@ -2305,7 +2305,7 @@
|
|
2305 |
"f":0.6502463054
|
2306 |
}
|
2307 |
},
|
2308 |
-
"speed":
|
2309 |
},
|
2310 |
"sources":[
|
2311 |
{
|
|
|
1 |
{
|
2 |
"lang":"pl",
|
3 |
"name":"core_news_md",
|
4 |
+
"version":"3.6.0",
|
5 |
"description":"Polish pipeline optimized for CPU. Components: tok2vec, morphologizer, parser, lemmatizer (trainable_lemmatizer), tagger, senter, ner.",
|
6 |
"author":"Explosion",
|
7 |
"email":"contact@explosion.ai",
|
8 |
"url":"https://explosion.ai",
|
9 |
"license":"GNU GPL 3.0",
|
10 |
+
"spacy_version":">=3.6.0,<3.7.0",
|
11 |
+
"spacy_git_version":"cb4fdc83e",
|
12 |
"vectors":{
|
13 |
"width":300,
|
14 |
"vectors":20000,
|
|
|
2305 |
"f":0.6502463054
|
2306 |
}
|
2307 |
},
|
2308 |
+
"speed":7821.6133862982
|
2309 |
},
|
2310 |
"sources":[
|
2311 |
{
|
morphologizer/cfg
CHANGED
@@ -1,5 +1,6 @@
|
|
1 |
{
|
2 |
"extend":false,
|
|
|
3 |
"labels_morph":{
|
4 |
"Animacy=Hum|Case=Nom|Gender=Masc|Number=Sing|POS=NOUN":"Animacy=Hum|Case=Nom|Gender=Masc|Number=Sing",
|
5 |
"AdpType=Prep|POS=ADP|Variant=Short":"AdpType=Prep|Variant=Short",
|
|
|
1 |
{
|
2 |
"extend":false,
|
3 |
+
"label_smoothing":0.0,
|
4 |
"labels_morph":{
|
5 |
"Animacy=Hum|Case=Nom|Gender=Masc|Number=Sing|POS=NOUN":"Animacy=Hum|Case=Nom|Gender=Masc|Number=Sing",
|
6 |
"AdpType=Prep|POS=ADP|Variant=Short":"AdpType=Prep|Variant=Short",
|
pl_core_news_md-any-py3-none-any.whl
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 sha256:68637c95d9d8380ea4eb6ee5e719df5fdcc2a7061f1cce1112de6101513fc2cc
|
3 |
+
size 49491414
|
tagger/cfg
CHANGED
@@ -1,4 +1,5 @@
|
|
1 |
{
|
|
|
2 |
"labels":[
|
3 |
"ADJ",
|
4 |
"ADJA",
|
|
|
1 |
{
|
2 |
+
"label_smoothing":0.0,
|
3 |
"labels":[
|
4 |
"ADJ",
|
5 |
"ADJA",
|