metadata

language: ar
license: apache-2.0
library_name: peft
base_model: CAMeL-Lab/bert-base-arabic-camelbert-mix
tags:
  - arabic
  - dialect-classification
  - lora

HammaLoRACAMeLBert

Advanced Arabic Dialect Classification Model with Complete Training Metrics

Full Training History

epoch	train_loss	eval_loss	train_accuracy	eval_accuracy	f1	precision	recall
1	2.0154	1.87199	0.315684	0.314607	0.278725	0.273851	0.315684
2	1.4809	1.09191	0.617445	0.629775	0.595584	0.620764	0.617445
3	1.0622	0.928158	0.679133	0.687079	0.676854	0.687955	0.679133
4	0.9443	0.82644	0.714286	0.711236	0.710725	0.716035	0.714286
5	0.8663	0.753623	0.745754	0.740449	0.746578	0.751243	0.745754
6	0.811	0.710841	0.763299	0.751685	0.764064	0.771564	0.763299
7	0.7637	0.661208	0.77741	0.76573	0.778244	0.782777	0.77741
8	0.7277	0.636298	0.783904	0.770225	0.785828	0.794191	0.783904
9	0.7061	0.616007	0.789461	0.769101	0.791083	0.797592	0.789461
10	0.6889	0.594658	0.798264	0.775843	0.799	0.802585	0.798264
11	0.6729	0.58317	0.801823	0.783146	0.802991	0.807269	0.801823
12	0.6591	0.58294	0.801886	0.780337	0.803151	0.809606	0.801886
13	0.6515	0.570984	0.807255	0.782022	0.808294	0.812656	0.807255
14	0.6435	0.563709	0.809441	0.783146	0.81018	0.813134	0.809441
15	0.64	0.562957	0.808816	0.783708	0.809795	0.813538	0.808816

Label Mapping:

{0: 'Egypt', 1: 'Iraq', 2: 'Lebanon', 3: 'Morocco', 4: 'Saudi_Arabia', 5: 'Sudan', 6: 'Tunisia'}

USAGE Example:

from transformers import pipeline

classifier = pipeline(
    "text-classification",
    model="Hamma-16/HammaLoRACAMeLBert",
    device="cuda" if torch.cuda.is_available() else "cpu"
)

sample_text = "شلونك اليوم؟"
result = classifier(sample_text)
print(f"Text: {sample_text}")
print(f"Predicted: {result[0]['label']} (confidence: {result[0]['score']:.1%})")