Views
No views yet
1from transformers import RobertaTokenizer, RobertaForSequenceClassification
2import torch
3import json
4from huggingface_hub import hf_hub_download
5
6# Load model and tokenizer
7model = RobertaForSequenceClassification.from_pretrained("vincenzoooooo/saskia-sonja-frida-conscientiousness")
8tokenizer = RobertaTokenizer.from_pretrained("vincenzoooooo/saskia-sonja-frida-conscientiousness")
9
10# Load label encoder
11label_encoder_path = hf_hub_download(repo_id="vincenzoooooo/saskia-sonja-frida-conscientiousness", filename="label_encoder.json")
12with open(label_encoder_path, 'r') as f:
13 label_data = json.load(f)
14 classes = label_data['classes'] # ['low', 'medium', 'high']
15
16# Make prediction
17text = "I love meeting new people and trying new experiences!"
18inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True, max_length=128)
19outputs = model(**inputs)
20predicted_class_id = torch.argmax(outputs.logits, dim=-1).item()
21prediction = classes[predicted_class_id]
22print(f"Conscientiousness: {prediction}")1@misc{saskia_sonja_frida_conscientiousness_2025,
2 title={Saskia, Sonja & Frida - Personality Detection System: Conscientiousness Prediction},
3 author={Saskia, Sonja & Frida},
4 year={2025},
5 howpublished={\url{https://huggingface.co/vincenzoooooo/saskia-sonja-frida-conscientiousness}},
6 note={NLP Shared Task 2025 - University of Antwerp}
7}