Update README.md
Browse files
README.md
CHANGED
@@ -25,7 +25,28 @@ The dataset comprises ~5M data points from three Latin American protest events:
|
|
25 |
|
26 |
## Example of Classification
|
27 |
|
28 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
29 |
|
30 |
## Validation Metrics
|
31 |
|
|
|
25 |
|
26 |
## Example of Classification
|
27 |
|
28 |
+
```python
|
29 |
+
## Pipeline as a high-level helper
|
30 |
+
from transformers import pipeline
|
31 |
+
toxic_classifier = pipeline("text-classification", model="bgonzalezbustamante/ft-xlm-roberta-toxicity")
|
32 |
+
|
33 |
+
## Non-toxic example
|
34 |
+
non_toxic = toxic_classifier("Que tengas un excelente día :)")
|
35 |
+
|
36 |
+
## Toxic example
|
37 |
+
toxic = toxic_classifier("Eres un maldito infeliz")
|
38 |
+
|
39 |
+
## Print examples
|
40 |
+
print(non_toxic)
|
41 |
+
print(toxic)
|
42 |
+
```
|
43 |
+
|
44 |
+
Output:
|
45 |
+
|
46 |
+
```
|
47 |
+
[{'label': 'NONTOXIC', 'score': 0.5529471635818481}]
|
48 |
+
[{'label': 'TOXIC', 'score': 0.6219274401664734}]
|
49 |
+
```
|
50 |
|
51 |
## Validation Metrics
|
52 |
|