mt5-small-finetuned-1feb-2

This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.3856
• Rouge1: 8.74
• Rouge2: 2.66
• Rougel: 8.58
• Rougelsum: 8.6

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
5.3153	1.0	311	2.6946	7.56	2.04	7.55	7.46
3.3159	2.0	622	2.5923	8.07	2.28	8.05	8.02
3.092	3.0	933	2.5342	7.83	2.01	7.81	7.76
2.9676	4.0	1244	2.4982	8.45	2.49	8.37	8.39
2.862	5.0	1555	2.4627	8.3	2.5	8.26	8.27
2.7891	6.0	1866	2.4366	8.67	2.81	8.53	8.55
2.7391	7.0	2177	2.4215	8.51	2.54	8.45	8.42
2.6887	8.0	2488	2.4277	8.71	2.53	8.56	8.54
2.6392	9.0	2799	2.3939	8.49	2.53	8.4	8.4
2.6139	10.0	3110	2.4015	9.28	2.85	9.14	9.19
2.5727	11.0	3421	2.3956	9.24	2.9	9.08	9.09
2.5595	12.0	3732	2.3856	8.45	2.59	8.31	8.35
2.5471	13.0	4043	2.3891	8.64	2.79	8.53	8.52
2.5231	14.0	4354	2.3870	8.78	2.79	8.64	8.6
2.5024	15.0	4665	2.3856	8.74	2.66	8.58	8.6

Framework versions

• Transformers 4.26.0
• Pytorch 1.13.1+cu116
• Datasets 2.9.0
• Tokenizers 0.13.2