commit files to HF hub

README.md

@@ -31,33 +31,33 @@ model-index:
     metrics:
     - name: BLEU4 (Question Answering)
       type: bleu4_question_answering
-      value: 5.66
+      value: 29.71
     - name: ROUGE-L (Question Answering)
       type: rouge_l_question_answering
-      value: 18.23
+      value: 55.07
     - name: METEOR (Question Answering)
       type: meteor_question_answering
-      value: 13.89
+      value: 41.65
     - name: BERTScore (Question Answering)
       type: bertscore_question_answering
-      value: 86.03
+      value: 94.96
     - name: MoverScore (Question Answering)
       type: moverscore_question_answering
-      value: 64.35
+      value: 83.99
     - name: AnswerF1Score (Question Answering)
       type: answer_f1_score__question_answering
-      value: 23.1
+      value: 73.33
     - name: AnswerExactMatch (Question Answering)
       type: answer_exact_match_question_answering
-      value: 5.52
+      value: 51.37
 ---
 
 # Model Card of `vocabtrimmer/mt5-small-trimmed-ru-120000-ruquad-qa`
-This model is fine-tuned version of [vocabtrimmer/mt5-small-trimmed-ru-120000](https://huggingface.co/vocabtrimmer/mt5-small-trimmed-ru-120000) for question answering task on the [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) (dataset_name: default) via [`lmqg`](https://github.com/asahi417/lm-question-generation).
+This model is fine-tuned version of [ckpts/mt5-small-trimmed-ru-120000](https://huggingface.co/ckpts/mt5-small-trimmed-ru-120000) for question answering task on the [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) (dataset_name: default) via [`lmqg`](https://github.com/asahi417/lm-question-generation).
 
 
 ### Overview
-- **Language model:** [vocabtrimmer/mt5-small-trimmed-ru-120000](https://huggingface.co/vocabtrimmer/mt5-small-trimmed-ru-120000)   
+- **Language model:** [ckpts/mt5-small-trimmed-ru-120000](https://huggingface.co/ckpts/mt5-small-trimmed-ru-120000)   
 - **Language:** ru  
 - **Training data:** [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) (default)
 - **Online Demo:** [https://autoqg.net/](https://autoqg.net/)
@@ -93,16 +93,16 @@ output = pipe("question: чем соответствует абсолютная
 
 |                  |   Score | Type    | Dataset                                                          |
 |:-----------------|--------:|:--------|:-----------------------------------------------------------------|
-| AnswerExactMatch |    5.52 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
-| AnswerF1Score    |   23.1  | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
-| BERTScore        |   86.03 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
-| Bleu_1           |   13.04 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
-| Bleu_2           |    9.76 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
-| Bleu_3           |    7.47 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
-| Bleu_4           |    5.66 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
-| METEOR           |   13.89 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
-| MoverScore       |   64.35 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
-| ROUGE_L          |   18.23 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
+| AnswerExactMatch |   51.37 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
+| AnswerF1Score    |   73.33 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
+| BERTScore        |   94.96 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
+| Bleu_1           |   46.17 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
+| Bleu_2           |   40.21 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
+| Bleu_3           |   34.84 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
+| Bleu_4           |   29.71 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
+| METEOR           |   41.65 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
+| MoverScore       |   83.99 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
+| ROUGE_L          |   55.07 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) |
 
 
 
@@ -114,12 +114,12 @@ The following hyperparameters were used during fine-tuning:
  - input_types: ['paragraph_question']
  - output_types: ['answer']
  - prefix_types: None
- - model: vocabtrimmer/mt5-small-trimmed-ru-120000
+ - model: ckpts/mt5-small-trimmed-ru-120000
  - max_length: 512
  - max_length_output: 32
- - epoch: 5
+ - epoch: 15
  - batch: 32
- - lr: 0.0005
+ - lr: 0.001
  - fp16: False
  - random_seed: 1
  - gradient_accumulation_steps: 2

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

@@ -1 +1 @@
-{"validation": {"Bleu_1": 0.13580593536793847, "Bleu_2": 0.10180848153479834, "Bleu_3": 0.078104486616958, "Bleu_4": 0.05876059281210598, "METEOR": 0.14077606116271782, "ROUGE_L": 0.1878528155634676, "BERTScore": 0.862298309187552, "MoverScore": 0.6441214873855503, "AnswerF1Score": 23.783942259933223, "AnswerExactMatch": 5.083399523431295}, "test": {"Bleu_1": 0.13035002966352713, "Bleu_2": 0.09760592518211351, "Bleu_3": 0.07470465677821343, "Bleu_4": 0.05661204561255439, "METEOR": 0.13894982525240993, "ROUGE_L": 0.18225659441453088, "BERTScore": 0.8602602633575867, "MoverScore": 0.6435025334899285, "AnswerF1Score": 23.104944192723003, "AnswerExactMatch": 5.520254169976171}}
+{"validation": {"Bleu_1": 0.48334806955494575, "Bleu_2": 0.42446150972912017, "Bleu_3": 0.37188475189985354, "Bleu_4": 0.32157021715601214, "METEOR": 0.42513802327329575, "ROUGE_L": 0.5669332394759723, "BERTScore": 0.953661481582806, "MoverScore": 0.8476297354040059, "AnswerF1Score": 75.26023674768254, "AnswerExactMatch": 53.4154090548054}, "test": {"Bleu_1": 0.4617225885020593, "Bleu_2": 0.40205062666505914, "Bleu_3": 0.34837286365826947, "Bleu_4": 0.2971343256502367, "METEOR": 0.4164766087640628, "ROUGE_L": 0.5506870736401595, "BERTScore": 0.9496106096470138, "MoverScore": 0.8398631419298846, "AnswerF1Score": 73.32765776541284, "AnswerExactMatch": 51.37013502779984}}