Hugging Face's logo

lmqg
/
mt5-small-ruquad-qg-ae

Text2Text Generation
Transformers
PyTorch
Russian
mt5
question generation
answer extraction
Eval Results
Inference Endpoints
Model card Files Community
1
asahi417 commited on
Commit
a9d4e4b
•
1 Parent(s): 8bddb9f

model update

Browse files
Files changed (1) hide show
  1. README.md +7 -7
README.md CHANGED
@@ -53,22 +53,22 @@ model-index:
53
  value: 0.6501855498621383
54
  - name: QAAlignedF1Score (BERTScore)
55
  type: qa_aligned_f1_score_bertscore
56
- value: 0.800455479300929
57
  - name: QAAlignedRecall (BERTScore)
58
  type: qa_aligned_recall_bertscore
59
- value: 0.8458137798797499
60
  - name: QAAlignedPrecision (BERTScore)
61
  type: qa_aligned_precision_bertscore
62
- value: 0.7609828008807118
63
  - name: QAAlignedF1Score (MoverScore)
64
  type: qa_aligned_f1_score_moverscore
65
- value: 0.569028191710343
66
  - name: QAAlignedRecall (MoverScore)
67
  type: qa_aligned_recall_moverscore
68
- value: 0.6040612143980683
69
  - name: QAAlignedPrecision (MoverScore)
70
  type: qa_aligned_precision_moverscore
71
- value: 0.5401911557620577
72
  ---
73
 
74
  # Model Card of `lmqg/mt5-small-ruquad-multitask`
@@ -141,7 +141,7 @@ question = pipe('generate question: Нелишним будет отметить
141
 
142
  | Dataset | Type | QA Aligned F1 Score (BERTScore) | QA Aligned F1 Score (MoverScore) | Link |
143
  |:--------|:-----|--------------------------------:|---------------------------------:|-----:|
144
- | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | default | 0.8 | 0.569 | [link](https://huggingface.co/lmqg/mt5-small-ruquad-multitask/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qg_ruquad.default.json) |
145
 
146
 
147
 
 
53
  value: 0.6501855498621383
54
  - name: QAAlignedF1Score (BERTScore)
55
  type: qa_aligned_f1_score_bertscore
56
+ value: 0.7973821321725308
57
  - name: QAAlignedRecall (BERTScore)
58
  type: qa_aligned_recall_bertscore
59
+ value: 0.8383368717732346
60
  - name: QAAlignedPrecision (BERTScore)
61
  type: qa_aligned_precision_bertscore
62
+ value: 0.7615392588632003
63
  - name: QAAlignedF1Score (MoverScore)
64
  type: qa_aligned_f1_score_moverscore
65
+ value: 0.5668638524057164
66
  - name: QAAlignedRecall (MoverScore)
67
  type: qa_aligned_recall_moverscore
68
+ value: 0.5979098322311048
69
  - name: QAAlignedPrecision (MoverScore)
70
  type: qa_aligned_precision_moverscore
71
+ value: 0.5410917350743871
72
  ---
73
 
74
  # Model Card of `lmqg/mt5-small-ruquad-multitask`
 
141
 
142
  | Dataset | Type | QA Aligned F1 Score (BERTScore) | QA Aligned F1 Score (MoverScore) | Link |
143
  |:--------|:-----|--------------------------------:|---------------------------------:|-----:|
144
+ | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | default | 0.797 | 0.567 | [link](https://huggingface.co/lmqg/mt5-small-ruquad-multitask/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qg_ruquad.default.json) |
145
 
146
 
147