Update README.md
Browse files
README.md
CHANGED
@@ -13,27 +13,27 @@ widget:
|
|
13 |
A fine-tuned model for fact-classification task on English tweets about COVID-19/vaccine.
|
14 |
|
15 |
## Intended uses & limitations
|
|
|
|
|
16 |
|
17 |
#### How to use
|
18 |
-
|
|
|
|
|
19 |
```python
|
20 |
from transformers import pipeline
|
21 |
pipe = pipeline("sentiment-analysis", model = "ans/vaccinating-covid-tweets")
|
22 |
-
seq =
|
23 |
-
|
24 |
pipe(seq)
|
25 |
```
|
26 |
|
27 |
-
Expected output
|
28 |
-
```
|
29 |
-
|
30 |
-
pipe(seq)
|
31 |
```
|
32 |
-
[{'label': 'true', 'score': 0.9996959567070007}]
|
33 |
|
34 |
#### Limitations and bias
|
35 |
-
|
36 |
-
Provide examples of latent issues and potential remediations.
|
37 |
|
38 |
## Training data & Procedure
|
39 |
|
@@ -62,7 +62,10 @@ Provide examples of latent issues and potential remediations.
|
|
62 |
- Misleading: misleading, exaggerated, out of context, needs context
|
63 |
- True: true, correct
|
64 |
|
65 |
-
##
|
|
|
|
|
|
|
66 |
|
67 |
# Contributors
|
68 |
- This page is a part of final team project from MLDL for DS class at SNU
|
|
|
13 |
A fine-tuned model for fact-classification task on English tweets about COVID-19/vaccine.
|
14 |
|
15 |
## Intended uses & limitations
|
16 |
+
You can classify if the input tweet (or any others statement) about COVID-19/vaccine is true, false or misleading.
|
17 |
+
Note that since this model was trained with data up to May 2020, the most recent information may not be reflected.
|
18 |
|
19 |
#### How to use
|
20 |
+
You can use this model directly on this page or using `transformers` in python.
|
21 |
+
|
22 |
+
- Load pipeline and implement with input sequence
|
23 |
```python
|
24 |
from transformers import pipeline
|
25 |
pipe = pipeline("sentiment-analysis", model = "ans/vaccinating-covid-tweets")
|
26 |
+
seq = "COVID-19 vaccines are safe and effective."
|
|
|
27 |
pipe(seq)
|
28 |
```
|
29 |
|
30 |
+
- Expected output
|
31 |
+
```python
|
32 |
+
[{'label': 'true', 'score': 0.9987803101539612}]
|
|
|
33 |
```
|
|
|
34 |
|
35 |
#### Limitations and bias
|
36 |
+
To conservatively classify whether an input sequence is true or not, the model may have predictions biased toward false/misleading.
|
|
|
37 |
|
38 |
## Training data & Procedure
|
39 |
|
|
|
62 |
- Misleading: misleading, exaggerated, out of context, needs context
|
63 |
- True: true, correct
|
64 |
|
65 |
+
## Evaluation results
|
66 |
+
| Training loss | Validation loss | Training accuracy | Validation accuracy |
|
67 |
+
| --- | --- | --- | --- |
|
68 |
+
| 0.1062 | 0.1006 | 96.3% | 94.5% |
|
69 |
|
70 |
# Contributors
|
71 |
- This page is a part of final team project from MLDL for DS class at SNU
|