asahi417 commited on
Commit
ac4a214
1 Parent(s): 2626bf4

model update

Browse files
Files changed (1) hide show
  1. README.md +102 -15
README.md CHANGED
@@ -1,24 +1,111 @@
 
1
  ---
2
- language:
3
- - en
4
- tags:
5
- - question generation
6
- - question answer generation
7
- license: mit
8
- datasets:
9
- - squad
10
  metrics:
11
- - bleu
12
  - meteor
13
- - rouge
 
 
 
 
 
 
 
 
14
  widget:
15
  - text: "<hl> Beyonce <hl> further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records."
16
- example_title: "Example 1"
17
  - text: "Beyonce further expanded her acting career, starring as blues singer <hl> Etta James <hl> in the 2008 musical biopic, Cadillac Records."
18
- example_title: "Example 2"
19
  - text: "Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, <hl> Cadillac Records <hl> ."
20
- example_title: "Example 3"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21
  ---
22
 
23
- # T5 finetuned on Question Generation
24
- T5 model for question generation. Please visit [our repository](https://github.com/asahi417/t5-question-generation) for more detail.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
  ---
3
+ license: cc-by-4.0
 
 
 
 
 
 
 
4
  metrics:
5
+ - bleu4
6
  - meteor
7
+ - rouge-l
8
+ - bertscore
9
+ - moverscore
10
+ language: en
11
+ datasets:
12
+ - lmqg/qg_squad
13
+ pipeline_tag: text2text-generation
14
+ tags:
15
+ - question generation
16
  widget:
17
  - text: "<hl> Beyonce <hl> further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records."
18
+ example_title: "Question Generation Example 1"
19
  - text: "Beyonce further expanded her acting career, starring as blues singer <hl> Etta James <hl> in the 2008 musical biopic, Cadillac Records."
20
+ example_title: "Question Generation Example 2"
21
  - text: "Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, <hl> Cadillac Records <hl> ."
22
+ example_title: "Question Generation Example 3"
23
+ model-index:
24
+ - name: lmqg/bart-base-squad-default
25
+ results:
26
+ - task:
27
+ name: Text2text Generation
28
+ type: text2text-generation
29
+ dataset:
30
+ name: lmqg/qg_squad
31
+ type: default
32
+ args: default
33
+ metrics:
34
+ - name: BLEU4
35
+ type: bleu4
36
+ value: 0.24400683975723286
37
+ - name: ROUGE-L
38
+ type: rouge-l
39
+ value: 0.5248717762722197
40
+ - name: METEOR
41
+ type: meteor
42
+ value: 0.25985179074924764
43
+ - name: BERTScore
44
+ type: bertscore
45
+ value: 0.9086545639846674
46
+ - name: MoverScore
47
+ type: moverscore
48
+ value: 0.644778282518877
49
  ---
50
 
51
+ # Language Models Fine-tuning on Question Generation: `lmqg/bart-base-squad-default`
52
+ This model is fine-tuned version of [facebook/bart-base](https://huggingface.co/facebook/bart-base) for question generation task on the
53
+ [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) (dataset_name: default).
54
+
55
+ ### Overview
56
+ - **Language model:** [facebook/bart-base](https://huggingface.co/facebook/bart-base)
57
+ - **Language:** en
58
+ - **Training data:** [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) (default)
59
+ - **Online Demo:** [https://autoqg.net/](https://autoqg.net/)
60
+ - **Repository:** [https://github.com/asahi417/lm-question-generation](https://github.com/asahi417/lm-question-generation)
61
+ - **Paper:** [TBA](TBA)
62
+
63
+ ### Usage
64
+ ```python
65
+
66
+ from transformers import pipeline
67
+
68
+ model_path = 'lmqg/bart-base-squad-default'
69
+ pipe = pipeline("text2text-generation", model_path)
70
+
71
+ # Question Generation
72
+ input_text = '<hl> Beyonce <hl> further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records.'
73
+ generation = pipe(input_text)
74
+ ```
75
+
76
+ ## Evaluation Metrics
77
+ ### In-domain Metrics
78
+
79
+
80
+ | Dataset | Type | BLEU4 | ROUGE-L | METEOR | BERTScore | MoverScore | Link |
81
+ |:--------|:-----|------:|--------:|-------:|----------:|-----------:|-----:|
82
+ | [lmqg/qg_squad](https://huggingface.co/datasets/lmqg/qg_squad) | default | 0.24400683975723286 | 0.5248717762722197 | 0.25985179074924764 | 0.9086545639846674 | 0.644778282518877 | [link](https://huggingface.co/lmqg/bart-base-squad-default/raw/main/eval/metric.first.sentence.paragraph_answer.question.lmqg_qg_squad.default.json) |
83
+
84
+
85
+ ### Out-of-domain Metrics
86
+
87
+
88
+
89
+ ## Training hyperparameters
90
+
91
+ The following hyperparameters were used during fine-tuning:
92
+ - dataset_path: lmqg/qg_squad
93
+ - dataset_name: default
94
+ - input_types: ['paragraph_answer']
95
+ - output_types: ['question']
96
+ - prefix_types: None
97
+ - model: facebook/bart-base
98
+ - max_length: 512
99
+ - max_length_output: 32
100
+ - epoch: 10
101
+ - batch: 32
102
+ - lr: 1.25e-05
103
+ - fp16: False
104
+ - random_seed: 1
105
+ - gradient_accumulation_steps: 1
106
+ - label_smoothing: 0.1
107
+
108
+ The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/bart-base-squad-default/raw/main/trainer_config.json).
109
+
110
+ ## Citation
111
+ TBA