Update README.md

README.md

@@ -28,7 +28,15 @@ model-index:
     - name: Morph (UFeats) Accuracy
       type: accuracy
       value: 0.8943839061
+license: cc-by-sa-4.0
+metrics:
+- accuracy
+- f1
+pipeline_tag: token-classification
 ---
+
+### SpaCy model for Classical Armenian trained on the UD_Classical_Armenian-CAVaL corpus
+
 | Feature | Description |
 | --- | --- |
 | **Name** | `xcl_ud_caval` |
@@ -37,9 +45,29 @@ model-index:
 | **Default Pipeline** | `tok2vec`, `morphologizer`, `tagger` |
 | **Components** | `tok2vec`, `morphologizer`, `tagger` |
 | **Vectors** | 0 keys, 0 unique vectors (0 dimensions) |
-| **Sources** | n/a |
-| **License** | n/a |
-| **Author** | [n/a]() |
+| **Training Data:** | Trained on: [UD_Classical_Armenian-CAVaL](https://github.com/UniversalDependencies/UD_Classical_Armenian-CAVaL), additional unreleased dataset|
+| **License** | CC BY-SA 4.0 |
+| **Author** | [Lilit Kharatyan](https://github.com/LilitKharatyan), [Petr Kocharov](https://github.com/pkocharov)|
+
+## Custom Language Setup
+
+To use this model with the custom Classical Armenian language class (`XclLanguage`), run the following initialization script before loading the model:
+
+```python
+from spacy.language import Language as DefaultLanguage
+from spacy.util import set_lang_class
+
+class XclLanguage(DefaultLanguage):
+    lang = "xcl"
+
+    @classmethod
+    def create(cls, vocab):
+        nlp = super(XclLanguage, cls).create(vocab)
+        return nlp
+
+# Add the custom language class to SpaCy
+set_lang_class("xcl", XclLanguage)
+```
 
 ### Label Scheme