model update

README.md

@@ -30,6 +30,36 @@ model-index:
     - name: BLEU4 (Question & Answer Generation)
       type: bleu4_question_answer_generation
       value: 0.87
+    - name: ROUGE-L (Question & Answer Generation)
+      type: rouge_l_question_answer_generation
+      value: 8.97
+    - name: METEOR (Question & Answer Generation)
+      type: meteor_question_answer_generation
+      value: 14.25
+    - name: BERTScore (Question & Answer Generation)
+      type: bertscore_question_answer_generation
+      value: 59.63
+    - name: MoverScore (Question & Answer Generation)
+      type: moverscore_question_answer_generation
+      value: 59.15
+    - name: QAAlignedF1Score-BERTScore (Question & Answer Generation)
+      type: qa_aligned_f1_score_bertscore_question_answer_generation
+      value: 76.88
+    - name: QAAlignedRecall-BERTScore (Question & Answer Generation)
+      type: qa_aligned_recall_bertscore_question_answer_generation
+      value: 76.69
+    - name: QAAlignedPrecision-BERTScore (Question & Answer Generation)
+      type: qa_aligned_precision_bertscore_question_answer_generation
+      value: 77.1
+    - name: QAAlignedF1Score-MoverScore (Question & Answer Generation)
+      type: qa_aligned_f1_score_moverscore_question_answer_generation
+      value: 77.95
+    - name: QAAlignedRecall-MoverScore (Question & Answer Generation)
+      type: qa_aligned_recall_moverscore_question_answer_generation
+      value: 77.66
+    - name: QAAlignedPrecision-MoverScore (Question & Answer Generation)
+      type: qa_aligned_precision_moverscore_question_answer_generation
+      value: 78.29
 ---
 
 # Model Card of `lmqg/mt5-base-koquad-qag`
@@ -71,12 +101,22 @@ output = pipe("1990년 영화 《 남부군 》에서 단역으로 영화배우
 
 - ***Metric (Question & Answer Generation)***: [raw metric file](https://huggingface.co/lmqg/mt5-base-koquad-qag/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qag_koquad.default.json) 
 
-|        |   Score | Type    | Dataset                                                            |
-|:-------|--------:|:--------|:-------------------------------------------------------------------|
-| Bleu_1 |    4.66 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
-| Bleu_2 |    2.43 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
-| Bleu_3 |    1.4  | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
-| Bleu_4 |    0.87 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+|                                 |   Score | Type    | Dataset                                                            |
+|:--------------------------------|--------:|:--------|:-------------------------------------------------------------------|
+| BERTScore                       |   59.63 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| Bleu_1                          |    4.66 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| Bleu_2                          |    2.43 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| Bleu_3                          |    1.4  | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| Bleu_4                          |    0.87 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| METEOR                          |   14.25 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| MoverScore                      |   59.15 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| QAAlignedF1Score (BERTScore)    |   76.88 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| QAAlignedF1Score (MoverScore)   |   77.95 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| QAAlignedPrecision (BERTScore)  |   77.1  | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| QAAlignedPrecision (MoverScore) |   78.29 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| QAAlignedRecall (BERTScore)     |   76.69 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| QAAlignedRecall (MoverScore)    |   77.66 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
+| ROUGE_L                         |    8.97 | default | [lmqg/qag_koquad](https://huggingface.co/datasets/lmqg/qag_koquad) |
 
 
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "lmqg_output/mt5-base-koquad-qag/best_model",
+  "_name_or_path": "lmqg_output/mt5-base-koquad-qag/model_umlauj/epoch_17",
   "add_prefix": false,
   "architectures": [
     "MT5ForConditionalGeneration"

eval/metric.first.answer.paragraph.questions_answers.lmqg_qag_koquad.default.json

@@ -1 +1 @@
-{"validation": {"Bleu_1": 0.22944459267164455, "Bleu_2": 0.14581809736110674, "Bleu_3": 0.09073762379558614, "Bleu_4": 0.060320148625246546}, "test": {"Bleu_1": 0.046633951117251785, "Bleu_2": 0.024267387176477355, "Bleu_3": 0.013983981210269144, "Bleu_4": 0.00865229909206871}}
+{"validation": {"Bleu_1": 0.22944459267164455, "Bleu_2": 0.14581809736110674, "Bleu_3": 0.09073762379558614, "Bleu_4": 0.060320148625246546, "METEOR": 0.21020536565003278, "ROUGE_L": 0.2466824670559609, "BERTScore": 0.7206163909596701, "MoverScore": 0.6696985735571672, "QAAlignedF1Score (BERTScore)": 0.7926473005991794, "QAAlignedRecall (BERTScore)": 0.7655120301171079, "QAAlignedPrecision (BERTScore)": 0.8227984495823168, "QAAlignedF1Score (MoverScore)": 0.8196813358976065, "QAAlignedRecall (MoverScore)": 0.7834570086693475, "QAAlignedPrecision (MoverScore)": 0.8610160754337751}, "test": {"Bleu_1": 0.046633951117251785, "Bleu_2": 0.024267387176477355, "Bleu_3": 0.013983981210269144, "Bleu_4": 0.00865229909206871, "METEOR": 0.14250042873681426, "ROUGE_L": 0.08972020374341763, "BERTScore": 0.5963447060618836, "MoverScore": 0.5914871388294957, "QAAlignedF1Score (BERTScore)": 0.7688378008310351, "QAAlignedRecall (BERTScore)": 0.7669275914942397, "QAAlignedPrecision (BERTScore)": 0.7709911078538925, "QAAlignedF1Score (MoverScore)": 0.7794858817862599, "QAAlignedRecall (MoverScore)": 0.7766280538644278, "QAAlignedPrecision (MoverScore)": 0.782938458908736}}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:46a4890a9a91302704ead27a34485781d792e07002252d411dfbecbfa2651a5e
-size 2329630901
+oid sha256:2c2530294b7f85ad1e4245396fa49b1484ecf584ccf9c1f971b983f3dad9c522
+size 2329634869

tokenizer_config.json

@@ -2,7 +2,7 @@
   "additional_special_tokens": null,
   "eos_token": "</s>",
   "extra_ids": 0,
-  "name_or_path": "lmqg_output/mt5-base-koquad-qag/best_model",
+  "name_or_path": "lmqg_output/mt5-base-koquad-qag/model_umlauj/epoch_17",
   "pad_token": "<pad>",
   "sp_model_kwargs": {},
   "special_tokens_map_file": "/home/patrick/.cache/torch/transformers/685ac0ca8568ec593a48b61b0a3c272beee9bc194a3c7241d15dcadb5f875e53.f76030f3ec1b96a8199b2593390c610e76ca8028ef3d24680000619ffb646276",