Update README.md
Browse files
README.md
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
datasets:
|
3 |
+
- squad
|
4 |
+
tags:
|
5 |
+
- question-generation
|
6 |
+
- distilt5
|
7 |
+
- distilt5-qg
|
8 |
+
widget:
|
9 |
+
- text: "generate question: <hl> 42 <hl> is the answer to life, the universe and everything. </s>"
|
10 |
+
- text: "question: What is 42 context: 42 is the answer to life, the universe and everything. </s>"
|
11 |
+
license: "MIT"
|
12 |
+
---
|
13 |
+
|
14 |
+
## DistilT5 for question-generation
|
15 |
+
This is distilled version of [t5-small-qa-qg-hl](https://huggingface.co/valhalla/t5-small-qa-qg-hl) model trained for question answering and answer aware question generation tasks.
|
16 |
+
|
17 |
+
The model is distilled using the **No Teacher Distillation** method proposed by Huggingface, [here](https://github.com/huggingface/transformers/tree/master/examples/seq2seq#distilbart).
|
18 |
+
|
19 |
+
We just copy alternating layers from `t5-small-qa-qg-hl` and finetune more on the same data. Following table lists other distilled models and their metrics.
|
20 |
+
|
21 |
+
| Name | BLEU-4 | METEOR | ROUGE-L | QA-EM | QA-F1 |
|
22 |
+
|---------------------------------------------------------------------------------|---------|---------|---------|--------|--------|
|
23 |
+
| [distilt5-qg-hl-6-4](https://huggingface.co/valhalla/distilt5-qg-hl-6-4) | 18.4141 | 24.8417 | 40.3435 | - | - |
|
24 |
+
| [distilt5-qa-qg-hl-6-4](https://huggingface.co/valhalla/distilt5-qa-qg-hl-6-4) | 18.6493 | 24.9685 | 40.5605 | 76.13 | 84.659 |
|
25 |
+
| [distilt5-qg-hl-12-6](https://huggingface.co/valhalla/distilt5-qg-hl-12-6) | 20.5275 | 26.5010 | 43.2676 | - | - |
|
26 |
+
| [distilt5-qa-qg-hl-12-6](https://huggingface.co/valhalla/distilt5-qa-qg-hl-12-6)| 20.6109 | 26.4533 | 43.0895 | 81.61 | 89.831 |
|
27 |
+
|
28 |
+
|
29 |
+
You can play with the model using the inference API. Here's how you can use it
|
30 |
+
|
31 |
+
For QG
|
32 |
+
|
33 |
+
`generate question: <hl> 42 <hl> is the answer to life, the universe and everything.`
|
34 |
+
|
35 |
+
For QA
|
36 |
+
|
37 |
+
`question: What is 42 context: 42 is the answer to life, the universe and everything.`
|
38 |
+
|
39 |
+
For more deatils see [this](https://github.com/patil-suraj/question_generation) repo.
|
40 |
+
|
41 |
+
### Model in action 🚀
|
42 |
+
|
43 |
+
You'll need to clone the [repo](https://github.com/patil-suraj/question_generation).
|
44 |
+
|
45 |
+
[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/patil-suraj/question_generation/blob/master/question_generation.ipynb)
|
46 |
+
|
47 |
+
|
48 |
+
```python3
|
49 |
+
from pipelines import pipeline
|
50 |
+
nlp = pipeline("multitask-qa-qg", model="valhalla/distilt5-qa-qg-hl-6-4")
|
51 |
+
|
52 |
+
# to generate questions simply pass the text
|
53 |
+
nlp("42 is the answer to life, the universe and everything.")
|
54 |
+
=> [{'answer': '42', 'question': 'What is the answer to life?'}]
|
55 |
+
|
56 |
+
# for qa pass a dict with "question" and "context"
|
57 |
+
nlp({
|
58 |
+
"question": "What is 42 ?",
|
59 |
+
"context": "42 is the answer to life, the universe and everything."
|
60 |
+
})
|
61 |
+
=> 'the answer to life, the universe and everything'
|
62 |
+
```
|