Update spaCy pipeline

README.md

@@ -24,8 +24,8 @@ model-index:
 | Feature | Description |
 | --- | --- |
 | **Name** | `en_degree_ner` |
-| **Version** | `0.1.0` |
-| **spaCy** | `>=3.7.4,<3.8.0` |
+| **Version** | `0.2.0` |
+| **spaCy** | `>=3.6.1,<3.7.0` |
 | **Default Pipeline** | `tok2vec`, `ner` |
 | **Components** | `tok2vec`, `ner` |
 | **Vectors** | 0 keys, 0 unique vectors (0 dimensions) |

config.cfg

@@ -1,6 +1,6 @@
 [paths]
-train = "degree_ner_train.spacy"
-dev = "degree_ner_valid.spacy"
+train = "training_data/degree_ner_train.spacy"
+dev = "training_data/degree_ner_valid.spacy"
 vectors = "en_core_web_sm"
 init_tok2vec = null
 
@@ -17,7 +17,6 @@ before_creation = null
 after_creation = null
 after_pipeline_creation = null
 tokenizer = {"@tokenizers":"spacy.Tokenizer.v1"}
-vectors = {"@vectors":"spacy.Vectors.v1"}
 
 [components]
 
@@ -136,10 +135,14 @@ ents_per_type = null
 vectors = ${paths.vectors}
 init_tok2vec = ${paths.init_tok2vec}
 vocab_data = null
-lookups = null
 before_init = null
 after_init = null
 
 [initialize.components]
 
+[initialize.lookups]
+@misc = "spacy.LookupsDataLoader.v1"
+lang = ${nlp.lang}
+tables = ["lexeme_norm"]
+
 [initialize.tokenizer]

en_degree_ner-any-py3-none-any.whl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae21d2e7ab6d2eecc862390da6e56ea659758825b15781aae89c40790489e47c
-size 31925456
+oid sha256:ef8317dc8d7607ed3afa375f804c9ee93d74961af4d73308fb76977b8205d044
+size 31996372

meta.json

@@ -1,14 +1,14 @@
 {
   "lang":"en",
   "name":"degree_ner",
-  "version":"0.1.0",
+  "version":"0.2.0",
   "description":"",
   "author":"",
   "email":"",
   "url":"",
   "license":"",
-  "spacy_version":">=3.7.4,<3.8.0",
-  "spacy_git_version":"bff8725f4",
+  "spacy_version":">=3.6.1,<3.7.0",
+  "spacy_git_version":"458bc5f45",
   "vectors":{
     "width":0,
     "vectors":0,
@@ -45,8 +45,8 @@
         "f":1.0
       }
     },
-    "tok2vec_loss":0.0000000273,
-    "ner_loss":0.000000224
+    "tok2vec_loss":0.0,
+    "ner_loss":0.0
   },
   "requirements":[
 

ner/moves

@@ -1 +1 @@
-��moves�p{"0":{},"1":{"DEGREE":280503},"2":{"DEGREE":280503},"3":{"DEGREE":280503},"4":{"DEGREE":280503,"":1},"5":{"":1}}�cfg��neg_key�
+��moves�p{"0":{},"1":{"DEGREE":484792},"2":{"DEGREE":484792},"3":{"DEGREE":484792},"4":{"DEGREE":484792,"":1},"5":{"":1}}�cfg��neg_key�

tok2vec/model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c47bcf6c1c21504e5f6276f591159d92526567eab0a7141cb684edf864f86f11
+oid sha256:a07b207de049fd5a5a2c38017156daa5b2c6110ca56870557b88ab26352aa973
 size 34126801

vocab/lookups.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76be8b528d0075f7aae98d6fa57a6d3c83ae480a8469e668d7b0af968995ac71
-size 1
+oid sha256:fce9c883c56165f29573cc938c2a1c9d417ac61bd8f56b671dd5f7996de70682
+size 70040

vocab/strings.json

@@ -12194,6 +12194,7 @@
   "Dionysius",
   "Diotrephes",
   "Dip",
+  "Dipl",
   "Diploma",
   "Diplomacy",
   "Diplomatic",
@@ -16111,6 +16112,7 @@
   "Higgenbotham",
   "Higgins",
   "High",
+  "HighSchool",
   "Higher",
   "Highest",
   "Highland",
@@ -16237,6 +16239,7 @@
   "Holcomb",
   "Hold",
   "Holden",
+  "Holder",
   "Holders",
   "Holding",
   "Holdings",
@@ -16322,6 +16325,7 @@
   "Honorable",
   "Honorary",
   "Honors",
+  "Honours",
   "Hood",
   "Hooker",
   "Hooks",
@@ -17939,6 +17943,7 @@
   "K",
   "K's",
   "K.",
+  "K12",
   "KAISER",
   "KB",
   "KC",
@@ -23387,6 +23392,7 @@
   "Pharpar",
   "Phase",
   "Phasing",
+  "Phd",
   "Phelan",
   "Phelps",
   "Phenix",
@@ -24066,6 +24072,7 @@
   "Prussia",
   "Pryce",
   "Pryor",
+  "Ps",
   "Psa",
   "Psalm",
   "Psalms",
@@ -44267,6 +44274,7 @@
   "dioxins",
   "dip",
   "diphtheria",
+  "dipl",
   "diploma",
   "diplomacy",
   "diplomas",
@@ -53000,6 +53008,7 @@
   "honour",
   "honourable",
   "honoured",
+  "honours",
   "hoo",
   "hood",
   "hooded",
@@ -55328,6 +55337,7 @@
   "ipe",
   "iph",
   "iphone",
+  "ipl",
   "ipo",
   "ipod",
   "ipods",
@@ -56401,6 +56411,7 @@
   "k.",
   "k.a",
   "[email protected]",
+  "k12",
   "kV",
   "kWh",
   "ka",