results_arat5-2_wiki

This model is a fine-tuned version of UBC-NLP/AraT5v2-base-1024 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 5.6421
• Rouge1: 0.0905
• Rouge2: 0.0
• Rougel: 0.0915
• Rougelsum: 0.0912
• Gen Len: 19.0

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
8.3962	0.9506	500	7.0927	0.0	0.0	0.0	0.0	0.0
7.072	1.9011	1000	7.0704	0.0	0.0	0.0	0.0	0.0
7.0441	2.8517	1500	7.0627	0.0	0.0	0.0	0.0	0.0
7.0044	3.8023	2000	7.0205	0.0	0.0	0.0	0.0	16.9719
6.9461	4.7529	2500	6.8398	0.0896	0.0	0.0908	0.0904	17.7903
6.727	5.7034	3000	6.5676	0.0905	0.0	0.0915	0.0912	18.8221
6.446	6.6540	3500	6.3711	0.0905	0.0	0.0915	0.0912	18.8221
6.3054	7.6046	4000	5.9586	0.0905	0.0	0.0915	0.0912	18.8933
5.8985	8.5551	4500	5.7386	0.0905	0.0	0.0915	0.0912	19.0
5.8333	9.5057	5000	5.6421	0.0905	0.0	0.0915	0.0912	19.0

Framework versions

• Transformers 4.42.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1