This model is fine-tuned for emotion recognition in Moroccan Darija text.
The architecture is based on DarijaBERT and is designed to classify user messages into emotion categories and emotional intensity levels.
The model was trained on a Moroccan Darija emotion dataset collected from social media conversations and manually annotated.
1from transformers import AutoTokenizer, AutoModelForSequenceClassification
2import torch
3
4model_name = "USERNAME/MODEL_NAME"
5
6tokenizer = AutoTokenizer.from_pretrained(model_name)
7model = AutoModelForSequenceClassification.from_pretrained(model_name)
8
9text = "كنحس براسي فرحان بزاف اليوم"
10
11inputs = tokenizer(
12 text,
13 return_tensors="pt",
14 truncation=True,
15 max_length=128
16)
17
18with torch.no_grad():
19 outputs = model(**inputs)
20
21prediction = torch.argmax(outputs.logits, dim=-1)
22
23print(prediction.item())
1@article{elguareh2026darijaemotion,
2 title={Emotion Recognition in Moroccan Darija using Transformer-Based Models},
3 author={El Guareh, Ayat Allah and others},
4 year={2026}
5}
