Create README.md

README.md

@@ -0,0 +1,52 @@
+---
+license: cc-by-4.0
+language:
+- ko
+- en
+base_model:
+- facebook/nllb-200-1.3B
+pipeline_tag: translation
+library_name: transformers
+tags:
+- pytorch
+- flores-200
+- Medical
+---
+
+* Explanation !
+  
+- This model is a fine-tuned version of the NLLB-200-1.3B model, specifically adapted for the medical terminology domain. All usage guidelines and copyright policies comply with those of the base model.
+
+- The fine-tuning dataset consists of the KMA Medical Terminology Collection and the KCD-8 masterfile's Korean-English description dataset.
+
+- It is specialized for translating Korean medical terms into English. ( ! Especially fitted for translating cause-of-death Korean text into English terms ! )
+
+- After pushing the model, we have continuously identified mistranslations and are updating the # Woondsc/nllb-1.3B-KMA-KCD-FFTtest (this model !)# model to address these issues. This model is an improved fine-tuned version specifically designed to correct additional mistranslations in the original model.
+
+- If you are looking to build a general Korean-to-English translation model for other purposes, feel free to use # Woondsc/nllb-1.3B-KMA-KCD # model. However, if you need better performance for Korean-to-English medical translations, we recommend using # Woondsc/nllb-1.3B-KMA-KCD-FFTtest (this model !)# instead.
+
+
+
+
+# Here is the example of using this model for translating Korean COD into English term . . .
+
+```python
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+
+# Load model directly
+tokenizer = AutoTokenizer.from_pretrained("Woondsc/nllb-1.3B-KMA-KCD-FFTtest")
+model = AutoModelForSeq2SeqLM.from_pretrained("Woondsc/nllb-1.3B-KMA-KCD-FFTtest")
+
+# Transformer function setting
+def translate(text, model, tokenizer, target_lang="eng_Latn"):
+    inputs = tokenizer(text, return_tensors="pt")
+    inputs["forced_bos_token_id"] = tokenizer.convert_tokens_to_ids(target_lang)
+    translated_tokens = model.generate(**inputs)
+    translated_text = tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)[0]
+    return translated_text
+
+# Execute example
+korean_text = "간질"
+english_translation = translate(korean_text, model, tokenizer)
+print("번역 결과:", english_translation)
+```