Create README.md

README.md

@@ -0,0 +1,30 @@
+# Model description
+
+- Morphosyntactic analyzer: Trankit
+- Tagset: UD
+- Embedding vectors: XLM-RoBERTa-Large
+- Dataset: NLPrePL-NKJP-fair-by-name (https://huggingface.co/datasets/ipipan/nlprepl)
+
+# How to use
+
+## Clone
+
+```
+git clone [email protected]:ipipan/nlpre_trankit_ud_xlm-roberta-large_nkjp-by-name
+```
+
+## Load model
+
+```
+import trankit
+
+model_path = './nlpre_trankit_ud_xlm-roberta-large_nkjp-by-name'
+
+trankit.verify_customized_pipeline(
+    category='customized-mwt', # pipeline category
+    save_dir=model_path, # directory used for saving models in previous steps
+    embedding_name='xlm-roberta-large' # embedding version that we use for training our customized pipeline, by default, it is `xlm-roberta-base`
+)
+
+model = trankit.Pipeline(lang='customized-mwt', cache_dir=model_path, embedding='xlm-roberta-large')
+```