Update README.md
Browse files
README.md
CHANGED
@@ -50,6 +50,27 @@ from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
|
|
50 |
|
51 |
tokenizer = AutoTokenizer.from_pretrained("tarudesu/ViHateT5-base-HSD")
|
52 |
model = AutoModelForSeq2SeqLM.from_pretrained("tarudesu/ViHateT5-base-HSD")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
53 |
```
|
54 |
|
55 |
Please feel free to contact us by email luannt@uit.edu.vn if you have any further information!
|
|
|
50 |
|
51 |
tokenizer = AutoTokenizer.from_pretrained("tarudesu/ViHateT5-base-HSD")
|
52 |
model = AutoModelForSeq2SeqLM.from_pretrained("tarudesu/ViHateT5-base-HSD")
|
53 |
+
|
54 |
+
def generate_output(input_text, prefix):
|
55 |
+
# Add prefix
|
56 |
+
prefixed_input_text = prefix + ': ' + input_text
|
57 |
+
|
58 |
+
# Tokenize input text
|
59 |
+
input_ids = tokenizer.encode(prefixed_input_text, return_tensors="pt")
|
60 |
+
|
61 |
+
# Generate output
|
62 |
+
output_ids = model.generate(input_ids, max_length=256)
|
63 |
+
|
64 |
+
# Decode the generated output
|
65 |
+
output_text = tokenizer.decode(output_ids[0], skip_special_tokens=True)
|
66 |
+
|
67 |
+
return output_text
|
68 |
+
|
69 |
+
sample = 'Tôi ghét bạn vl luôn!'
|
70 |
+
prefix = 'hate-spans-detection' # Choose 1 from 3 prefixes ['hate-speech-detection', 'toxic-speech-detection', 'hate-spans-detection']
|
71 |
+
|
72 |
+
result = generate_output(sample, prefix)
|
73 |
+
print('Result: ', result)
|
74 |
```
|
75 |
|
76 |
Please feel free to contact us by email luannt@uit.edu.vn if you have any further information!
|