model update

README.md

@@ -31,19 +31,19 @@ model-index:
     metrics:
     - name: BLEU4
       type: bleu4
-      value: 0.33315782056958804
+      value: 33.3157820569588
     - name: ROUGE-L
       type: rouge-l
-      value: 0.6223954899626488
+      value: 62.23954899626488
     - name: METEOR
       type: meteor
-      value: 0.35426819977024687
+      value: 35.426819977024685
     - name: BERTScore
       type: bertscore
-      value: 0.9458259980280204
+      value: 94.58259980280204
     - name: MoverScore
       type: moverscore
-      value: 0.8006984647429997
+      value: 80.06984647429996
     - name: AnswerF1Score (Question Answering)
       type: answer_f1_score_question_answering
       value: 69.3966081647612
@@ -93,18 +93,20 @@ output = pipe('question: What is a person called is practicing heresy?, context:
 
 ### Metric (Question Answering)
 
-|                  |     Score | Type    | Dataset                                                            | Link                                                                                                                                                    |
-|:-----------------|----------:|:--------|:-------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------|
-| AnswerExactMatch | 52.286    | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) | https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json |
-| AnswerF1Score    | 69.3966   | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) | https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json |
-| BERTScore        |  0.945826 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) | https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json |
-| Bleu_1           |  0.570705 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) | https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json |
-| Bleu_2           |  0.481677 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) | https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json |
-| Bleu_3           |  0.397816 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) | https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json |
-| Bleu_4           |  0.333158 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) | https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json |
-| METEOR           |  0.354268 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) | https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json |
-| MoverScore       |  0.800698 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) | https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json |
-| ROUGE_L          |  0.622395 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) | https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json |
+|                  |   Score | Type    | Dataset                                                            |
+|:-----------------|--------:|:--------|:-------------------------------------------------------------------|
+| AnswerExactMatch | 52.286  | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) |
+| AnswerF1Score    | 69.3966 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) |
+| BERTScore        | 94.5826 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) |
+| Bleu_1           | 57.0705 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) |
+| Bleu_2           | 48.1677 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) |
+| Bleu_3           | 39.7816 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) |
+| Bleu_4           | 33.3158 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) |
+| METEOR           | 35.4268 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) |
+| MoverScore       | 80.0698 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) |
+| ROUGE_L          | 62.2395 | default | [lmqg/qg_tweetqa](https://huggingface.co/datasets/lmqg/qg_tweetqa) |
+
+- [raw metric file](https://huggingface.co/lmqg/t5-base-tweetqa-question-answering/raw/main/eval/metric.first.answer.paragraph_question.answer.lmqg_qg_tweetqa.default.json)
 
 
 ## Training hyperparameters

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "lmqg_output/t5-base-tweetqa-question-answering/best_model",
+  "_name_or_path": "lmqg_output/t5-base-tweetqa-question-answering/model_eszyci/epoch_9",
   "add_prefix": false,
   "architectures": [
     "T5ForConditionalGeneration"

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:507ff75c34f5ca9b2a3ab171d05d7c573afdc8dbecfb71a3d48280b1940579cf
-size 891614207
+oid sha256:3d9af18f30d344085a0c0eb5fce63dbe9354c80444d241743054f9bc72089d2f
+size 891617855

tokenizer_config.json

@@ -104,7 +104,7 @@
   "eos_token": "</s>",
   "extra_ids": 100,
   "model_max_length": 512,
-  "name_or_path": "lmqg_output/t5-base-tweetqa-question-answering/best_model",
+  "name_or_path": "lmqg_output/t5-base-tweetqa-question-answering/model_eszyci/epoch_9",
   "pad_token": "<pad>",
   "special_tokens_map_file": null,
   "tokenizer_class": "T5Tokenizer",