commit files to HF hub

README.md

@@ -29,25 +29,25 @@ model-index:
     metrics:
     - name: BLEU4 (Question Answering)
       type: bleu4_question_answering
-      value: 5.52
+      value: 9.62
     - name: ROUGE-L (Question Answering)
       type: rouge_l_question_answering
-      value: 22.57
+      value: 30.92
     - name: METEOR (Question Answering)
       type: meteor_question_answering
-      value: 20.41
+      value: 26.47
     - name: BERTScore (Question Answering)
       type: bertscore_question_answering
-      value: 87.88
+      value: 90.14
     - name: MoverScore (Question Answering)
       type: moverscore_question_answering
-      value: 69.95
+      value: 74.5
     - name: AnswerF1Score (Question Answering)
       type: answer_f1_score__question_answering
-      value: 41.13
+      value: 51.47
     - name: AnswerExactMatch (Question Answering)
       type: answer_exact_match_question_answering
-      value: 29.1
+      value: 36.0
 ---
 
 # Model Card of `vocabtrimmer/mt5-small-trimmed-it-itquad-qa`
@@ -91,16 +91,16 @@ output = pipe("question: Quale batterio ha il nome del paese che colpisce di pi
 
 |                  |   Score | Type    | Dataset                                                          |
 |:-----------------|--------:|:--------|:-----------------------------------------------------------------|
-| AnswerExactMatch |   29.1  | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
-| AnswerF1Score    |   41.13 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
-| BERTScore        |   87.88 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
-| Bleu_1           |   12.54 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
-| Bleu_2           |    9.12 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
-| Bleu_3           |    7.07 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
-| Bleu_4           |    5.52 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
-| METEOR           |   20.41 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
-| MoverScore       |   69.95 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
-| ROUGE_L          |   22.57 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
+| AnswerExactMatch |   36    | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
+| AnswerF1Score    |   51.47 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
+| BERTScore        |   90.14 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
+| Bleu_1           |   20.22 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
+| Bleu_2           |   15.36 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
+| Bleu_3           |   12.16 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
+| Bleu_4           |    9.62 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
+| METEOR           |   26.47 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
+| MoverScore       |   74.5  | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
+| ROUGE_L          |   30.92 | default | [lmqg/qg_itquad](https://huggingface.co/datasets/lmqg/qg_itquad) |
 
 
 
@@ -115,12 +115,12 @@ The following hyperparameters were used during fine-tuning:
  - model: vocabtrimmer/mt5-small-trimmed-it
  - max_length: 512
  - max_length_output: 32
- - epoch: 24
+ - epoch: 13
  - batch: 32
- - lr: 0.0001
+ - lr: 0.0005
  - fp16: False
  - random_seed: 1
- - gradient_accumulation_steps: 4
+ - gradient_accumulation_steps: 2
  - label_smoothing: 0.15
 
 The full configuration can be found at [fine-tuning config file](https://huggingface.co/vocabtrimmer/mt5-small-trimmed-it-itquad-qa/raw/main/trainer_config.json).

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

@@ -1 +1 @@
-{"validation": {"Bleu_1": 0.13332586409724573, "Bleu_2": 0.0951085568992768, "Bleu_3": 0.07254502242763887, "Bleu_4": 0.05499781333956201, "METEOR": 0.22095922326981232, "ROUGE_L": 0.23618759920461654, "BERTScore": 0.890475623232257, "MoverScore": 0.7237123490207725, "AnswerF1Score": 45.447290367384184, "AnswerExactMatch": 34.130634774609014}, "test": {"Bleu_1": 0.12537137525474193, "Bleu_2": 0.09115582087333446, "Bleu_3": 0.07065444733813277, "Bleu_4": 0.05523335749426967, "METEOR": 0.2040968082393822, "ROUGE_L": 0.22574760175660866, "BERTScore": 0.8788482561141531, "MoverScore": 0.6995213256082121, "AnswerF1Score": 41.13401560531839, "AnswerExactMatch": 29.097121829412536}}
+{"validation": {"Bleu_1": 0.21083394217721843, "Bleu_2": 0.16030694672703755, "Bleu_3": 0.12768239875259021, "Bleu_4": 0.10107736671762473, "METEOR": 0.2817387368332181, "ROUGE_L": 0.3112476321282083, "BERTScore": 0.9142447584623857, "MoverScore": 0.7745192501933702, "AnswerF1Score": 55.97960751263587, "AnswerExactMatch": 42.98856617163885}, "test": {"Bleu_1": 0.2022175606832424, "Bleu_2": 0.15355997720884773, "Bleu_3": 0.12164802976426284, "Bleu_4": 0.09622641132646627, "METEOR": 0.2646909834754643, "ROUGE_L": 0.3092117112734317, "BERTScore": 0.9014100162690597, "MoverScore": 0.7449889153606168, "AnswerF1Score": 51.46633302083642, "AnswerExactMatch": 35.996845840452096}}