Views
No views yet
| Property | Value |
|---|---|
| Base model | aubmindlab/bert-base-arabertv2 |
| Language | Arabic (Modern Standard Arabic + some dialect) |
| Task | Text Classification |
| Number of classes | 89 |
| Training samples | ~800,000 |
| Evaluation metrics | Accuracy, F1 Macro, F1 Weighted |
| Hardware used | 1 × NVIDIA GPU (CUDA 12.8) |
| Framework | Transformers + PyTorch |
1from transformers import AutoTokenizer, AutoModelForSequenceClassification
2import torch
3import torch.nn.functional as F
4
5model_id = "YourUsername/arabic-medical-classifier-arabertv2"
6tokenizer = AutoTokenizer.from_pretrained(model_id)
7model = AutoModelForSequenceClassification.from_pretrained(model_id)
8model.eval()
9
10question = "ما هي أعراض ارتفاع ضغط الدم؟"
11answer = "من أهم الأعراض الصداع والدوخة والنزيف الأنفي أحيانًا."
12text = question + tokenizer.sep_token + answer
13
14inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True, max_length=512)
15with torch.no_grad():
16 outputs = model(**inputs)
17 probs = F.softmax(outputs.logits, dim=-1)
18 pred_id = torch.argmax(probs, dim=-1).item()
19
20print("Predicted class:", model.config.id2label[pred_id])
21print("Confidence:", probs[0][pred_id].item())