kompactss
/

JeBERT_je_ko

encoder-decoder

text2text-generation

Model card Files Files and versions

kompactss commited on May 2, 2022

Commit

430ba28

·

1 Parent(s): cee8623

Update README.md

Files changed (1) hide show

README.md +13 -13

README.md CHANGED Viewed

@@ -3,24 +3,24 @@ license: afl-3.0
 ---
 # 🍊 제주 방언 번역 모델 🍊
-  -  제주어 -> 표준어
-  -  Made by. 구름 자연어처리 과정 3기 3조!!
 ## 1. Seq2Seq Transformer Model
-  -  encoder :  Bert
-  -  decoder :  Bert
-  -  Tokenizer :  WordPiece Tokenizer
 ## 2. Dataset
-  -  Jit Dataset
-  -  AI HUB(+아래아 문자)
 ## 3. Hyper Parameters
-  -  Epoch :  10 epochs(best at 8 epoch)
-  -  Random Seed :  42
-  -  Learning Rate :  5e-5
-  -  Warm up Ratio :  0.1
-  -  Batch Size :  32
 ## 4. BLEU Score
-  -  Jit + AI HUB(+아래아 문자) Dataset :  79.0

 ---
 # 🍊 제주 방언 번역 모델 🍊
+    -  제주어 -> 표준어
+    -  Made by. 구름 자연어처리 과정 3기 3조!!
 ## 1. Seq2Seq Transformer Model
+    -  encoder :  Bert
+    -  decoder :  Bert
+    -  Tokenizer :  WordPiece Tokenizer
 ## 2. Dataset
+    -  Jit Dataset
+    -  AI HUB(+아래아 문자)
 ## 3. Hyper Parameters
+    -  Epoch :  10 epochs(best at 8 epoch)
+    -  Random Seed :  42
+    -  Learning Rate :  5e-5
+    -  Warm up Ratio :  0.1
+    -  Batch Size :  32
 ## 4. BLEU Score
+    -  Jit + AI HUB(+아래아 문자) Dataset :  79.0