asahi417 commited on
Commit
9ac105b
1 Parent(s): df82984

model update

Browse files
Files changed (26) hide show
  1. README.md +118 -0
  2. eval/{metric.first.answer.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.first.answer.paragraph_answer.question.lmqg_qg_frquad.default.json} +0 -0
  3. eval/{metric.first.answer.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.first.answer.paragraph_sentence.question.lmqg_qg_frquad.default.json} +0 -0
  4. eval/{metric.first.answer.sentence_answer.question.asahi417_qg_frquad.default.json → metric.first.answer.sentence_answer.question.lmqg_qg_frquad.default.json} +0 -0
  5. eval/{metric.first.sentence.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.first.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json} +0 -0
  6. eval/{metric.first.sentence.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.first.sentence.paragraph_sentence.question.lmqg_qg_frquad.default.json} +0 -0
  7. eval/{metric.first.sentence.sentence_answer.question.asahi417_qg_frquad.default.json → metric.first.sentence.sentence_answer.question.lmqg_qg_frquad.default.json} +0 -0
  8. eval/{metric.last.sentence.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.last.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json} +0 -0
  9. eval/{metric.last.sentence.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.last.sentence.paragraph_sentence.question.lmqg_qg_frquad.default.json} +0 -0
  10. eval/{metric.last.sentence.sentence_answer.question.asahi417_qg_frquad.default.json → metric.last.sentence.sentence_answer.question.lmqg_qg_frquad.default.json} +0 -0
  11. eval/{metric.long.sentence.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.long.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json} +0 -0
  12. eval/{metric.long.sentence.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.long.sentence.paragraph_sentence.question.lmqg_qg_frquad.default.json} +0 -0
  13. eval/{metric.long.sentence.sentence_answer.question.asahi417_qg_frquad.default.json → metric.long.sentence.sentence_answer.question.lmqg_qg_frquad.default.json} +0 -0
  14. eval/{metric.middle.sentence.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.middle.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json} +0 -0
  15. eval/{metric.middle.sentence.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.middle.sentence.paragraph_sentence.question.lmqg_qg_frquad.default.json} +0 -0
  16. eval/{metric.middle.sentence.sentence_answer.question.asahi417_qg_frquad.default.json → metric.middle.sentence.sentence_answer.question.lmqg_qg_frquad.default.json} +0 -0
  17. eval/{metric.short.sentence.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.short.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json} +0 -0
  18. eval/{metric.short.sentence.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.short.sentence.paragraph_sentence.question.lmqg_qg_frquad.default.json} +0 -0
  19. eval/{metric.short.sentence.sentence_answer.question.asahi417_qg_frquad.default.json → metric.short.sentence.sentence_answer.question.lmqg_qg_frquad.default.json} +0 -0
  20. eval/{samples.test.hyp.paragraph_answer.question.asahi417_qg_frquad.default.txt → samples.test.hyp.paragraph_answer.question.lmqg_qg_frquad.default.txt} +0 -0
  21. eval/{samples.test.hyp.paragraph_sentence.question.asahi417_qg_frquad.default.txt → samples.test.hyp.paragraph_sentence.question.lmqg_qg_frquad.default.txt} +0 -0
  22. eval/{samples.test.hyp.sentence_answer.question.asahi417_qg_frquad.default.txt → samples.test.hyp.sentence_answer.question.lmqg_qg_frquad.default.txt} +0 -0
  23. eval/{samples.validation.hyp.paragraph_answer.question.asahi417_qg_frquad.default.txt → samples.validation.hyp.paragraph_answer.question.lmqg_qg_frquad.default.txt} +0 -0
  24. eval/{samples.validation.hyp.paragraph_sentence.question.asahi417_qg_frquad.default.txt → samples.validation.hyp.paragraph_sentence.question.lmqg_qg_frquad.default.txt} +0 -0
  25. eval/{samples.validation.hyp.sentence_answer.question.asahi417_qg_frquad.default.txt → samples.validation.hyp.sentence_answer.question.lmqg_qg_frquad.default.txt} +0 -0
  26. trainer_config.json +1 -1
README.md ADDED
@@ -0,0 +1,118 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ ---
3
+ license: cc-by-4.0
4
+ metrics:
5
+ - bleu4
6
+ - meteor
7
+ - rouge-l
8
+ - bertscore
9
+ - moverscore
10
+ language: fr
11
+ datasets:
12
+ - lmqg/qg_frquad
13
+ pipeline_tag: text2text-generation
14
+ tags:
15
+ - question generation
16
+ - answer extraction
17
+ widget:
18
+ - text: "generate question: Créateur » (Maker), lui aussi au singulier, « <hl> le Suprême Berger <hl> » (The Great Shepherd) ; de l'autre, des réminiscences de la théologie de l'Antiquité : le tonnerre, voix de Jupiter, « Et souvent ta voix gronde en un tonnerre terrifiant », etc."
19
+ example_title: "Question Generation Example 1"
20
+ - text: "generate question: Ce black dog peut être lié à des évènements traumatisants issus du monde extérieur, tels que son renvoi de l'Amirauté après la catastrophe des Dardanelles, lors de la <hl> Grande Guerre <hl> de 14-18, ou son rejet par l'électorat en juillet 1945."
21
+ example_title: "Question Generation Example 2"
22
+ - text: "generate question: contre <hl> Normie Smith <hl> et 15 000 dollars le 28 novembre 1938."
23
+ example_title: "Question Generation Example 3"
24
+ - text: "Pourtant, la strophe spensérienne, utilisée cinq fois avant que ne commence le chœur, constitue en soi un vecteur dont les répétitions structurelles, selon Ricks, relèvent du pur lyrisme tout en constituant une menace potentielle. Après les huit sages pentamètres iambiques, l'alexandrin final <hl> permet une pause <hl>, « véritable illusion d'optique » qu'accentuent les nombreuses expressions archaïsantes telles que did swoon, did seem, did go, did receive, did make, qui doublent le prétérit en un temps composé et paraissent à la fois « très précautionneuses et très peu pressées »."
25
+ example_title: "Answer Extraction Example 1"
26
+ - text: "Néanmoins, une fois encore, l'arithmétique modulaire est insuffisante pour venir à bout du théorème. Dirichlet utilise de nombreuses techniques analytiques, comme les séries entières et l'analyse complexe. Le fruit de ces travaux donne naissance à une nouvelle branche des mathématiques : la théorie analytique des nombres. L'un des points cruciaux de cette théorie provient de l'unique article de <hl> Bernhard Riemann <hl> en théorie des nombres : Sur le nombre de nombres premiers inférieurs à une taille donnée. Il conjecture une localisation des racines de sa fonction ζ. La recherche de la position des racines, initiée par Dirichlet, devient une préoccupation centrale et reste l'une des conjectures pressenties comme les plus difficiles des mathématiques de notre époque."
27
+ example_title: "Answer Extraction Example 2"
28
+ model-index:
29
+ - name: lmqg/mt5-small-frquad-multitask
30
+ results:
31
+ - task:
32
+ name: Text2text Generation
33
+ type: text2text-generation
34
+ dataset:
35
+ name: lmqg/qg_frquad
36
+ type: default
37
+ args: default
38
+ metrics:
39
+ - name: BLEU4
40
+ type: bleu4
41
+ value: 0.0774587808669847
42
+ - name: ROUGE-L
43
+ type: rouge-l
44
+ value: 0.28060033794696104
45
+ - name: METEOR
46
+ type: meteor
47
+ value: 0.17623658069800158
48
+ - name: BERTScore
49
+ type: bertscore
50
+ value: 0.7990349418715752
51
+ - name: MoverScore
52
+ type: moverscore
53
+ value: 0.5644356433018314
54
+ ---
55
+
56
+ # Language Models Fine-tuning on Question Generation: `lmqg/mt5-small-frquad-multitask`
57
+ This model is fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) for question generation task on the
58
+ [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) (dataset_name: default).
59
+ This model is fine-tuned on the answer extraction task as well as the question generation.
60
+
61
+ ### Overview
62
+ - **Language model:** [google/mt5-small](https://huggingface.co/google/mt5-small)
63
+ - **Language:** fr
64
+ - **Training data:** [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) (default)
65
+ - **Online Demo:** [https://autoqg.net/](https://autoqg.net/)
66
+ - **Repository:** [https://github.com/asahi417/lm-question-generation](https://github.com/asahi417/lm-question-generation)
67
+ - **Paper:** [TBA](TBA)
68
+
69
+ ### Usage
70
+ ```python
71
+
72
+ from transformers import pipeline
73
+
74
+ model_path = 'lmqg/mt5-small-frquad-multitask'
75
+ pipe = pipeline("text2text-generation", model_path)
76
+
77
+ # Question Generation
78
+ input_text = 'generate question: Créateur » (Maker), lui aussi au singulier, « <hl> le Suprême Berger <hl> » (The Great Shepherd) ; de l'autre, des réminiscences de la théologie de l'Antiquité : le tonnerre, voix de Jupiter, « Et souvent ta voix gronde en un tonnerre terrifiant », etc.'
79
+ question = pipe(input_text)
80
+ # Answer Extraction
81
+ answer = pipe('extract answers: Pourtant, la strophe spensérienne, utilisée cinq fois avant que ne commence le chœur, constitue en soi un vecteur dont les répétitions structurelles, selon Ricks, relèvent du pur lyrisme tout en constituant une menace potentielle. Après les huit sages pentamètres iambiques, l'alexandrin final <hl> permet une pause <hl>, « véritable illusion d'optique » qu'accentuent les nombreuses expressions archaïsantes telles que did swoon, did seem, did go, did receive, did make, qui doublent le prétérit en un temps composé et paraissent à la fois « très précautionneuses et très peu pressées ».')
82
+ ```
83
+
84
+ ## Evaluation Metrics
85
+
86
+
87
+ ### Metrics
88
+
89
+ | Dataset | Type | BLEU4 | ROUGE-L | METEOR | BERTScore | MoverScore | Link |
90
+ |:--------|:-----|------:|--------:|-------:|----------:|-----------:|-----:|
91
+ | [lmqg/qg_frquad](https://huggingface.co/datasets/lmqg/qg_frquad) | default | 0.0774587808669847 | 0.28060033794696104 | 0.17623658069800158 | 0.7990349418715752 | 0.5644356433018314 | [link](https://huggingface.co/lmqg/mt5-small-frquad-multitask/raw/main/eval/metric.first.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json) |
92
+
93
+
94
+
95
+
96
+ ## Training hyperparameters
97
+
98
+ The following hyperparameters were used during fine-tuning:
99
+ - dataset_path: lmqg/qg_frquad
100
+ - dataset_name: default
101
+ - input_types: ['paragraph_answer', 'paragraph_sentence']
102
+ - output_types: ['question', 'answer']
103
+ - prefix_types: ['qg', 'ae']
104
+ - model: google/mt5-small
105
+ - max_length: 512
106
+ - max_length_output: 32
107
+ - epoch: 18
108
+ - batch: 64
109
+ - lr: 0.0005
110
+ - fp16: False
111
+ - random_seed: 1
112
+ - gradient_accumulation_steps: 1
113
+ - label_smoothing: 0.15
114
+
115
+ The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/mt5-small-frquad-multitask/raw/main/trainer_config.json).
116
+
117
+ ## Citation
118
+ TBA
eval/{metric.first.answer.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.first.answer.paragraph_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.first.answer.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.first.answer.paragraph_sentence.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.first.answer.sentence_answer.question.asahi417_qg_frquad.default.json → metric.first.answer.sentence_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.first.sentence.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.first.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.first.sentence.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.first.sentence.paragraph_sentence.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.first.sentence.sentence_answer.question.asahi417_qg_frquad.default.json → metric.first.sentence.sentence_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.last.sentence.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.last.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.last.sentence.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.last.sentence.paragraph_sentence.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.last.sentence.sentence_answer.question.asahi417_qg_frquad.default.json → metric.last.sentence.sentence_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.long.sentence.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.long.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.long.sentence.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.long.sentence.paragraph_sentence.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.long.sentence.sentence_answer.question.asahi417_qg_frquad.default.json → metric.long.sentence.sentence_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.middle.sentence.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.middle.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.middle.sentence.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.middle.sentence.paragraph_sentence.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.middle.sentence.sentence_answer.question.asahi417_qg_frquad.default.json → metric.middle.sentence.sentence_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.short.sentence.paragraph_answer.question.asahi417_qg_frquad.default.json → metric.short.sentence.paragraph_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.short.sentence.paragraph_sentence.question.asahi417_qg_frquad.default.json → metric.short.sentence.paragraph_sentence.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{metric.short.sentence.sentence_answer.question.asahi417_qg_frquad.default.json → metric.short.sentence.sentence_answer.question.lmqg_qg_frquad.default.json} RENAMED
File without changes
eval/{samples.test.hyp.paragraph_answer.question.asahi417_qg_frquad.default.txt → samples.test.hyp.paragraph_answer.question.lmqg_qg_frquad.default.txt} RENAMED
File without changes
eval/{samples.test.hyp.paragraph_sentence.question.asahi417_qg_frquad.default.txt → samples.test.hyp.paragraph_sentence.question.lmqg_qg_frquad.default.txt} RENAMED
File without changes
eval/{samples.test.hyp.sentence_answer.question.asahi417_qg_frquad.default.txt → samples.test.hyp.sentence_answer.question.lmqg_qg_frquad.default.txt} RENAMED
File without changes
eval/{samples.validation.hyp.paragraph_answer.question.asahi417_qg_frquad.default.txt → samples.validation.hyp.paragraph_answer.question.lmqg_qg_frquad.default.txt} RENAMED
File without changes
eval/{samples.validation.hyp.paragraph_sentence.question.asahi417_qg_frquad.default.txt → samples.validation.hyp.paragraph_sentence.question.lmqg_qg_frquad.default.txt} RENAMED
File without changes
eval/{samples.validation.hyp.sentence_answer.question.asahi417_qg_frquad.default.txt → samples.validation.hyp.sentence_answer.question.lmqg_qg_frquad.default.txt} RENAMED
File without changes
trainer_config.json CHANGED
@@ -1 +1 @@
1
- {"dataset_path": "asahi417/qg_frquad", "dataset_name": "default", "input_types": ["paragraph_answer", "paragraph_sentence"], "output_types": ["question", "answer"], "prefix_types": ["qg", "ae"], "model": "google/mt5-small", "max_length": 512, "max_length_output": 32, "epoch": 18, "batch": 64, "lr": 0.0005, "fp16": false, "random_seed": 1, "gradient_accumulation_steps": 1, "label_smoothing": 0.15}
 
1
+ {"dataset_path": "lmqg/qg_frquad", "dataset_name": "default", "input_types": ["paragraph_answer", "paragraph_sentence"], "output_types": ["question", "answer"], "prefix_types": ["qg", "ae"], "model": "google/mt5-small", "max_length": 512, "max_length_output": 32, "epoch": 18, "batch": 64, "lr": 0.0005, "fp16": false, "random_seed": 1, "gradient_accumulation_steps": 1, "label_smoothing": 0.15}