Update spaCy pipeline

README.md

@@ -78,8 +78,8 @@ Finnish pipeline optimized for CPU. Components: tok2vec, tagger, morphologizer,
 | Feature | Description |
 | --- | --- |
 | **Name** | `fi_core_news_lg` |
-| **Version** | `3.5.0` |
-| **spaCy** | `>=3.5.0,<3.6.0` |
+| **Version** | `3.6.0` |
+| **spaCy** | `>=3.6.0,<3.7.0` |
 | **Default Pipeline** | `tok2vec`, `tagger`, `morphologizer`, `parser`, `lemmatizer`, `attribute_ruler`, `ner` |
 | **Components** | `tok2vec`, `tagger`, `morphologizer`, `parser`, `lemmatizer`, `senter`, `attribute_ruler`, `ner` |
 | **Vectors** | floret (200000, 300) |

accuracy.json

@@ -378,11 +378,6 @@
       "r": 0.8258196721,
       "f": 0.83436853
     },
-    "WORK_OF_ART": {
-      "p": 0.6363636364,
-      "r": 0.375,
-      "f": 0.4719101124
-    },
     "CARDINAL": {
       "p": 0.9365079365,
       "r": 0.9407744875,
@@ -393,6 +388,11 @@
       "r": 0.7875288684,
       "f": 0.7830080367
     },
+    "WORK_OF_ART": {
+      "p": 0.6363636364,
+      "r": 0.375,
+      "f": 0.4719101124
+    },
     "PRODUCT": {
       "p": 0.7095435685,
       "r": 0.6951219512,
@@ -454,5 +454,5 @@
       "f": 0.8275862069
     }
   },
-  "speed": 7752.7325229573
+  "speed": 7690.3240081011
 }

config.cfg

@@ -46,6 +46,7 @@ upstream = "tok2vec"
 [components.morphologizer]
 factory = "morphologizer"
 extend = false
+label_smoothing = 0.0
 overwrite = true
 scorer = {"@scorers":"spacy.morphologizer_scorer.v1"}
 
@@ -143,6 +144,7 @@ maxout_pieces = 2
 
 [components.tagger]
 factory = "tagger"
+label_smoothing = 0.0
 neg_prefix = "!"
 overwrite = false
 scorer = {"@scorers":"spacy.tagger_scorer.v1"}

fi_core_news_lg-any-py3-none-any.whl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e781d20f357b5490e9fdc9c1e2340e7eedbc1f3a0cb050270ee0105229a75999
-size 230768203
+oid sha256:4f9ea84fc160d85ce185f664b099d49b909d7cf620af52a436a0290a920f7e85
+size 230768256

meta.json

@@ -1,14 +1,14 @@
 {
   "lang":"fi",
   "name":"core_news_lg",
-  "version":"3.5.0",
+  "version":"3.6.0",
   "description":"Finnish pipeline optimized for CPU. Components: tok2vec, tagger, morphologizer, parser, lemmatizer (trainable_lemmatizer), senter, ner.",
   "author":"Explosion",
   "email":"contact@explosion.ai",
   "url":"https://explosion.ai",
   "license":"CC BY-SA 4.0",
-  "spacy_version":">=3.5.0,<3.6.0",
-  "spacy_git_version":"9e0322de1",
+  "spacy_version":">=3.6.0,<3.7.0",
+  "spacy_git_version":"cb4fdc83e",
   "vectors":{
     "width":300,
     "vectors":200000,
@@ -2578,11 +2578,6 @@
         "r":0.8258196721,
         "f":0.83436853
       },
-      "WORK_OF_ART":{
-        "p":0.6363636364,
-        "r":0.375,
-        "f":0.4719101124
-      },
       "CARDINAL":{
         "p":0.9365079365,
         "r":0.9407744875,
@@ -2593,6 +2588,11 @@
         "r":0.7875288684,
         "f":0.7830080367
       },
+      "WORK_OF_ART":{
+        "p":0.6363636364,
+        "r":0.375,
+        "f":0.4719101124
+      },
       "PRODUCT":{
         "p":0.7095435685,
         "r":0.6951219512,
@@ -2654,7 +2654,7 @@
         "f":0.8275862069
       }
     },
-    "speed":7752.7325229573
+    "speed":7690.3240081011
   },
   "sources":[
     {

morphologizer/cfg

@@ -1,5 +1,6 @@
 {
   "extend":false,
+  "label_smoothing":0.0,
   "labels_morph":{
     "Case=Nom|Number=Sing|POS=NOUN":"Case=Nom|Number=Sing",
     "NumType=Ord|POS=ADJ":"NumType=Ord",

tagger/cfg

@@ -1,4 +1,5 @@
 {
+  "label_smoothing":0.0,
   "labels":[
     "A",
     "Adj",