t5-abs-2209-2133-lr-0.0001-bs-10-maxep-10

This model is a fine-tuned version of google-t5/t5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.1096
• Rouge/rouge1: 0.4008
• Rouge/rouge2: 0.1771
• Rouge/rougel: 0.3596
• Rouge/rougelsum: 0.3602
• Bertscore/bertscore-precision: 0.9032
• Bertscore/bertscore-recall: 0.8756
• Bertscore/bertscore-f1: 0.889
• Meteor: 0.3185
• Gen Len: 31.4

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: 0.0001
• train_batch_size: 10
• eval_batch_size: 10
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 20
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge/rouge1	Rouge/rouge2	Rouge/rougel	Rouge/rougelsum	Bertscore/bertscore-precision	Bertscore/bertscore-recall	Bertscore/bertscore-f1	Meteor	Gen Len
2.8028	0.8	2	2.1830	0.3216	0.1347	0.2803	0.2806	0.9067	0.8675	0.8865	0.2452	25.4
1.7773	2.0	5	2.1405	0.3287	0.1186	0.2839	0.2847	0.8962	0.8697	0.8825	0.2613	31.5
2.4733	2.8	7	2.1404	0.3409	0.1254	0.2934	0.294	0.8977	0.8725	0.8847	0.2714	32.7
1.6093	4.0	10	2.1299	0.4007	0.1538	0.3276	0.3302	0.9052	0.8761	0.8902	0.3121	33.0
2.3586	4.8	12	2.1212	0.4092	0.1834	0.3627	0.363	0.9067	0.8774	0.8917	0.3327	31.8
1.4823	6.0	15	2.1129	0.396	0.1745	0.3553	0.3567	0.9029	0.8762	0.8892	0.3163	32.0
2.1644	6.8	17	2.1106	0.3936	0.1731	0.3559	0.3574	0.903	0.8753	0.8888	0.3092	31.2
1.4411	8.0	20	2.1096	0.4008	0.1771	0.3596	0.3602	0.9032	0.8756	0.889	0.3185	31.4

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1