amharic-news-model

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.1030
• Accuracy: 0.7938
• Macro Precision: 0.7863
• Macro Recall: 0.7621
• Macro F1: 0.7704

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 1
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro Precision	Macro Recall	Macro F1
0.9303	1.0	451	0.6758	0.7732	0.7571	0.7571	0.7571
0.907	2.0	902	0.8922	0.8041	0.7908	0.7873	0.7890
0.6074	3.0	1353	1.2265	0.7629	0.7697	0.7090	0.7199
0.3266	4.0	1804	1.2719	0.7938	0.8071	0.7450	0.7587
0.1429	5.0	2255	1.2891	0.8041	0.8158	0.7589	0.7728

Framework versions

• Transformers 4.45.0
• Pytorch 2.10.0+cu128
• Datasets 4.6.1
• Tokenizers 0.20.3