t5-base-standardized-color

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

• Loss: 0.2702
• Rouge1: 58.8296
• Rouge2: 50.9332
• Rougel: 58.2604
• Rougelsum: 58.323
• Gen Len: 16.2521

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: 8
• eval_batch_size: 8
• 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
No log	1.0	236	0.3490	49.2479	40.2468	48.6246	48.5062	18.0148
No log	2.0	472	0.3080	52.8701	44.4405	52.3371	52.2684	17.1589
0.3775	3.0	708	0.2871	55.4404	46.9716	54.9257	54.8833	16.9004
0.3775	4.0	944	0.2792	61.4338	53.5456	60.9375	61.0613	15.0636
0.2834	5.0	1180	0.2789	56.7293	48.3876	56.1734	56.2194	16.6589
0.2834	6.0	1416	0.2742	53.2995	44.7666	52.7346	52.7591	17.3644
0.2553	7.0	1652	0.2757	57.3854	49.1456	56.6424	56.7503	16.5318
0.2553	8.0	1888	0.2717	56.9399	48.9799	56.405	56.4246	16.7055
0.2393	9.0	2124	0.2703	58.4279	50.4598	57.8832	57.9165	16.3856
0.2393	10.0	2360	0.2702	58.8296	50.9332	58.2604	58.323	16.2521

Framework versions

• Transformers 4.33.0
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3