Migrate model card from transformers-repo
Browse filesRead announcement at https://discuss.huggingface.co/t/announcement-all-model-cards-will-be-migrated-to-hf-co-model-repos/2755
Original file history: https://github.com/huggingface/transformers/commits/master/model_cards/NeuML/bert-small-cord19qa/README.md
README.md
ADDED
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# BERT-Small fine-tuned on CORD-19 QA dataset
|
2 |
+
|
3 |
+
[bert-small-cord19-squad model](https://huggingface.co/NeuML/bert-small-cord19-squad2) fine-tuned on the [CORD-19 QA dataset](https://www.kaggle.com/davidmezzetti/cord19-qa?select=cord19-qa.json).
|
4 |
+
|
5 |
+
## CORD-19 QA dataset
|
6 |
+
The CORD-19 QA dataset is a SQuAD 2.0 formatted list of question, context, answer combinations covering the [CORD-19 dataset](https://www.semanticscholar.org/cord19).
|
7 |
+
|
8 |
+
## Building the model
|
9 |
+
|
10 |
+
```bash
|
11 |
+
python run_squad.py \
|
12 |
+
--model_type bert \
|
13 |
+
--model_name_or_path bert-small-cord19-squad \
|
14 |
+
--do_train \
|
15 |
+
--do_lower_case \
|
16 |
+
--version_2_with_negative \
|
17 |
+
--train_file cord19-qa.json \
|
18 |
+
--per_gpu_train_batch_size 8 \
|
19 |
+
--learning_rate 5e-5 \
|
20 |
+
--num_train_epochs 10.0 \
|
21 |
+
--max_seq_length 384 \
|
22 |
+
--doc_stride 128 \
|
23 |
+
--output_dir bert-small-cord19qa \
|
24 |
+
--save_steps 0 \
|
25 |
+
--threads 8 \
|
26 |
+
--overwrite_cache \
|
27 |
+
--overwrite_output_dir
|
28 |
+
```
|
29 |
+
|
30 |
+
## Testing the model
|
31 |
+
|
32 |
+
Example usage below:
|
33 |
+
|
34 |
+
```python
|
35 |
+
from transformers import pipeline
|
36 |
+
|
37 |
+
qa = pipeline(
|
38 |
+
"question-answering",
|
39 |
+
model="NeuML/bert-small-cord19qa",
|
40 |
+
tokenizer="NeuML/bert-small-cord19qa"
|
41 |
+
)
|
42 |
+
|
43 |
+
qa({
|
44 |
+
"question": "What is the median incubation period?",
|
45 |
+
"context": "The incubation period is around 5 days (range: 4-7 days) with a maximum of 12-13 day"
|
46 |
+
})
|
47 |
+
|
48 |
+
qa({
|
49 |
+
"question": "What is the incubation period range?",
|
50 |
+
"context": "The incubation period is around 5 days (range: 4-7 days) with a maximum of 12-13 day"
|
51 |
+
})
|
52 |
+
|
53 |
+
qa({
|
54 |
+
"question": "What type of surfaces does it persist?",
|
55 |
+
"context": "The virus can survive on surfaces for up to 72 hours such as plastic and stainless steel ."
|
56 |
+
})
|
57 |
+
```
|
58 |
+
|
59 |
+
```json
|
60 |
+
{"score": 0.5970273583242793, "start": 32, "end": 38, "answer": "5 days"}
|
61 |
+
{"score": 0.999555868193891, "start": 39, "end": 56, "answer": "(range: 4-7 days)"}
|
62 |
+
{"score": 0.9992726505196998, "start": 61, "end": 88, "answer": "plastic and stainless steel"}
|
63 |
+
```
|