Update spaCy pipeline

README.md

@@ -13,18 +13,18 @@ model-index:
     metrics:
     - name: NER Precision
       type: precision
-      value: 0.8598145591
+      value: 0.909263658
     - name: NER Recall
       type: recall
-      value: 0.846751632
+      value: 0.8684210526
     - name: NER F Score
       type: f_score
-      value: 0.8532331004
+      value: 0.8883731724
 ---
 | Feature | Description |
 | --- | --- |
 | **Name** | `de_ner_d_dist` |
-| **Version** | `0.0.8` |
+| **Version** | `2.0.0` |
 | **spaCy** | `>=3.8.7,<3.9.0` |
 | **Default Pipeline** | `transformer`, `ner` |
 | **Components** | `transformer`, `ner` |
@@ -49,33 +49,6 @@ model-index:
 
 | Type | Score |
 | --- | --- |
-| `ENTS_F` | 85.13 |
-| `ENTS_P` | 86.37 |
-| `ENTS_R` | 83.91 |
-| `TRANSFORMER_LOSS` | 220810.34 |
-| `NER_LOSS` | 377981.31 |
-
-## Evaluation on Conll03 (updated version from 2006)
-| Type | Score | 
-| --- | --- |
-| `ENTS_F` | 81.34 |
-| `ENTS_P` | 85.76 |
-| `ENTS_R` | 77.35 |
-
-### Per Type
-| Type | F-Score | 
-| --- | --- |
-| `ORG` | 69.86 |
-| `LOC` | 79.14 |
-| `PER` | 93.53|
-| `MISC` | 37.17 |
-
-## Datasets used for training
-* D. Benikova, C. Biemann, M. Reznicek (2014). NoSta-D Named Entity Annotation for German: Guidelines and Dataset. Proceedings of LREC 2014, Reykjavik, Iceland.
-* M. Schiersch, V. Mironova, M. Schmitt, P. Thomas, A. Gabryszak, L. Hennig (2018). A German Corpus for Fine-Grained Named Entity Recognition and Relation Extraction of Traffic and Industry Events. Proceedings of LREC 2018, Miyazaki, Japan.
-* J. Zöllner, K. Sperfeld, C. Wick, R. Labahn (2021). Optimizing Small BERTs Trained for German NER. Information 2021, 12, 443.
-* M. Ehrmann, M. Romanello, A. Flückiger, and S. Clematide (2020). Extended Overview of CLEF HIPE 2020: Named Entity Processing on Historical Newspapers in Working Notes of CLEF 2020 - Conference and Labs of the Evaluation Forum, Thessaloniki, Greece, 2020, vol. 2696, p. 38. doi: 10.5281/zenodo.4117566.
-* L. Hennig, P. T. Truong, A. Gabryszak (2021). Mobie: A German Dataset for Named Entity Recognition, Entity Linking and Relation Extraction in the Mobility Domain. arXiv preprint arXiv:2108.06955.
-* A. Hamdi, E. Linhares Pontes, E. Boros, T. T. H. Nguyen, G. Hackl, J. G. Moreno, A. Doucet (2021). Multilingual Dataset for Named Entity Recognition, Entity Linking and Stance Detection in Historical Newspapers (V1.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.4573313
-* J. Nothman, N. Ringland, W. Radford, T. Murphy, J. R. Curran (2013). Learning Multilingual Named Entity Recognition from Wikipedia. Artificial Intelligence, 194, 151-175.
-* S. Schweter (2025). HisGermaNER (Revision 83571b3). doi: 10.57967/hf/5770, https://huggingface.co/datasets/stefan-it/HisGermaNER.
+| `ENTS_P` | 90.93 |
+| `ENTS_R` | 86.84 |
+| `ENTS_F` | 88.84 |

config.cfg

@@ -1,6 +1,6 @@
 [paths]
-train = "corpus/train.spacy"
-dev = "corpus/dev.spacy"
+train = "dataset/ner-d.train.spacy"
+dev = "dataset/ner-d.dev.spacy"
 vectors = null
 init_tok2vec = null
 

de_ner_d_dist-any-py3-none-any.whl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f7c1e26e65312c76089387006bd96c1defff2d355c476d0088e69d47b1cbb54
-size 407601001
+oid sha256:df1086959a4b0f37ca5e2c0eb04b59deb1ee9c42d974e915f4dca809a45d9dfa
+size 407768587

meta.json

@@ -1,7 +1,7 @@
 {
   "lang":"de",
   "name":"ner_d_dist",
-  "version":"0.0.8",
+  "version":"2.0.0",
   "description":"",
   "author":"",
   "email":"",
@@ -38,33 +38,36 @@
 
   ],
   "performance":{
-    "ents_f":0.8532331004,
-    "ents_p":0.8598145591,
-    "ents_r":0.846751632,
+    "token_acc":null,
+    "token_p":null,
+    "token_r":null,
+    "token_f":null,
+    "ents_p":0.909263658,
+    "ents_r":0.8684210526,
+    "ents_f":0.8883731724,
     "ents_per_type":{
-      "PER":{
-        "p":0.8720152818,
-        "r":0.9011350551,
-        "f":0.886336057
-      },
-      "MISC":{
-        "p":0.8093037585,
-        "r":0.7234921509,
-        "f":0.7639959285
+      "LOC":{
+        "p":0.9019607843,
+        "r":0.876747141,
+        "f":0.8891752577
       },
       "ORG":{
-        "p":0.8100463302,
-        "r":0.7707656613,
-        "f":0.7899179646
+        "p":0.8316582915,
+        "r":0.773364486,
+        "f":0.8014527845
       },
-      "LOC":{
-        "p":0.8846876067,
-        "r":0.8846876067,
-        "f":0.8846876067
+      "PER":{
+        "p":0.9554794521,
+        "r":0.9500567537,
+        "f":0.952760387
+      },
+      "MISC":{
+        "p":0.8484848485,
+        "r":0.5185185185,
+        "f":0.6436781609
       }
     },
-    "transformer_loss":1085.9911612031,
-    "ner_loss":1562.7839800269
+    "speed":28155.5503559723
   },
   "requirements":[
     "spacy-transformers>=1.3.9,<1.4.0"

ner/model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c305f2d865177cb86d7626a2686deb16e2736ec69619da6c19e372fc9de4e061
+oid sha256:c01412ea30d5595fce9344b6b3ae99c35f6d0299951c3b8cb65ebf34ba1f934d
 size 225962

ner/moves

@@ -1 +1 @@
-��moves��{"0":{},"1":{"LOC":123564,"PER":89901,"ORG":69200,"MISC":46310},"2":{"LOC":123564,"PER":89901,"ORG":69200,"MISC":46310},"3":{"LOC":123564,"PER":89901,"ORG":69200,"MISC":46310},"4":{"LOC":123564,"PER":89901,"ORG":69200,"MISC":46310,"":1},"5":{"":1}}�cfg��neg_key�
+��moves��{"0":{},"1":{"LOC":137997,"PER":96305,"ORG":74176,"MISC":49655},"2":{"LOC":137997,"PER":96305,"ORG":74176,"MISC":49655},"3":{"LOC":137997,"PER":96305,"ORG":74176,"MISC":49655},"4":{"LOC":137997,"PER":96305,"ORG":74176,"MISC":49655,"":1},"5":{"":1}}�cfg��neg_key�

transformer/model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9442cf82455edafc40feaf4f4ee239c9f72ef89cfefa01f70c4fafc503b85786
+oid sha256:990fd546cca7afed0c785f0f292e4e20990c3b7c49f8c3b6a4f5e7297f19ab92
 size 437388364