commit files to HF hub

README.md

@@ -31,25 +31,25 @@ model-index:
     metrics:
     - name: BLEU4 (Question Answering)
       type: bleu4_question_answering
-      value: 11.64
+      value: 26.33
     - name: ROUGE-L (Question Answering)
       type: rouge_l_question_answering
-      value: 25.15
+      value: 38.14
     - name: METEOR (Question Answering)
       type: meteor_question_answering
-      value: 18.54
+      value: 31.8
     - name: BERTScore (Question Answering)
       type: bertscore_question_answering
-      value: 87.94
+      value: 92.2
     - name: MoverScore (Question Answering)
       type: moverscore_question_answering
-      value: 67.71
+      value: 77.16
     - name: AnswerF1Score (Question Answering)
       type: answer_f1_score__question_answering
-      value: 36.62
+      value: 60.48
     - name: AnswerExactMatch (Question Answering)
       type: answer_exact_match_question_answering
-      value: 18.88
+      value: 39.34
 ---
 
 # Model Card of `lmqg/mbart-large-cc25-frquad-qa`
@@ -93,16 +93,16 @@ output = pipe("question: En quelle année a-t-on trouvé trace d'un haut fournea
 
 |                  |   Score | Type    | Dataset                                                          |
 |:-----------------|--------:|:--------|:-----------------------------------------------------------------|
-| AnswerExactMatch |   18.88 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
-| AnswerF1Score    |   36.62 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
-| BERTScore        |   87.94 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
-| Bleu_1           |   22.45 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
-| Bleu_2           |   17.59 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
-| Bleu_3           |   14.24 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
-| Bleu_4           |   11.64 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
-| METEOR           |   18.54 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
-| MoverScore       |   67.71 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
-| ROUGE_L          |   25.15 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
+| AnswerExactMatch |   39.34 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
+| AnswerF1Score    |   60.48 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
+| BERTScore        |   92.2  | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
+| Bleu_1           |   37.27 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
+| Bleu_2           |   32.61 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
+| Bleu_3           |   29.23 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
+| Bleu_4           |   26.33 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
+| METEOR           |   31.8  | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
+| MoverScore       |   77.16 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
+| ROUGE_L          |   38.14 | default | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) |
 
 
 
@@ -117,13 +117,13 @@ The following hyperparameters were used during fine-tuning:
  - model: facebook/mbart-large-cc25
  - max_length: 512
  - max_length_output: 32
- - epoch: 8
- - batch: 4
- - lr: 0.0005
+ - epoch: 15
+ - batch: 32
+ - lr: 0.0002
  - fp16: False
  - random_seed: 1
- - gradient_accumulation_steps: 16
- - label_smoothing: 0.0
+ - gradient_accumulation_steps: 2
+ - label_smoothing: 0.15
 
 The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/mbart-large-cc25-frquad-qa/raw/main/trainer_config.json).
 

eval/metric.first.answer.paragraph_question.answer.lmqg_qg_frquad.default.json

@@ -1 +1 @@
-{"validation": {"Bleu_1": 0.2486998514115745, "Bleu_2": 0.19685386062905924, "Bleu_3": 0.15927715488613817, "Bleu_4": 0.13079566202215717, "METEOR": 0.17776344222534668, "ROUGE_L": 0.2612935500550596, "BERTScore": 0.8770172560641278, "MoverScore": 0.6603900426723037, "AnswerF1Score": 36.315470546752344, "AnswerExactMatch": 13.582183186951067}, "test": {"Bleu_1": 0.22451698867420222, "Bleu_2": 0.17588940062174876, "Bleu_3": 0.14239704098395764, "Bleu_4": 0.11637303657440054, "METEOR": 0.18537576969401506, "ROUGE_L": 0.25151608670509873, "BERTScore": 0.879374634011776, "MoverScore": 0.6771203356899227, "AnswerF1Score": 36.6193739874486, "AnswerExactMatch": 18.88331242158093}}
+{"validation": {"Bleu_1": 0.4045265721243085, "Bleu_2": 0.35834231464728317, "Bleu_3": 0.3224689757382265, "Bleu_4": 0.2904498301545984, "AnswerF1Score": 60.947732333638, "AnswerExactMatch": 34.78670012547052, "METEOR": 0.29381562220907614, "ROUGE_L": 0.38980645663087565, "BERTScore": 0.9229774555118948, "MoverScore": 0.7561687416234889}, "test": {"Bleu_1": 0.37274341521157345, "Bleu_2": 0.3261271410730797, "Bleu_3": 0.2922922705608965, "Bleu_4": 0.2632611604554427, "AnswerF1Score": 60.483258743906056, "AnswerExactMatch": 39.33500627352572, "METEOR": 0.31795391099657566, "ROUGE_L": 0.3813527939689909, "BERTScore": 0.9220344831645862, "MoverScore": 0.77162094614282}}