|
--- |
|
language: |
|
- en |
|
metrics: |
|
- f1 |
|
pipeline_tag: text-classification |
|
--- |
|
|
|
|
|
# Initialize tokenizer and model |
|
tokenizer = BartTokenizer.from_pretrained('ihgn/paraphrase-detection') |
|
model = BartForConditionalGeneration.from_pretrained("ihgn/paraphrase-detection").to(device) |
|
source_sentence = "This was a series of nested angular standards , so that measurements in azimuth and elevation could be done directly in polar coordinates relative to the ecliptic." |
|
target_paraphrase = "This was a series of nested polar scales , so that measurements in azimuth and elevation could be performed directly in angular coordinates relative to the ecliptic" |
|
|
|
def paraphrase_detection(model, tokenizer, source_sentence, target_paraphrase): |
|
# Tokenize the input sentence |
|
inputs = tokenizer.encode_plus(source_sentence + ' <sep> ' + target_paraphrase, return_tensors='pt') |
|
|
|
# Classify the input using the model |
|
with torch.no_grad(): |
|
outputs = model.generate(inputs['input_ids'].to(device)) |
|
|
|
# Get the predicted label |
|
predicted_label = 1 if generated_text == '1' else 0 |
|
print("Predicted Label:", predicted_label) |
|
|
|
paraphrase_detection(model, tokenizer, source_sentence, target_paraphrase) |