mt5_deed_sum

This model is a fine-tuned version of google/mt5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5590
• Rouge1: 0.0
• Rouge2: 0.0
• Rougel: 0.0
• Rougelsum: 0.0
• 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
• lr_scheduler_warmup_steps: 5000
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
19.6644	1.0	188	13.7553	0.7175	0.0	0.7387	0.7032	12.283
10.6133	2.0	376	11.3168	0.7175	0.0	0.7387	0.7032	13.4654
10.1742	3.0	564	7.7772	0.7175	0.0	0.7387	0.7032	15.9874
2.8326	4.0	752	4.6011	0.7498	0.0	0.7731	0.7387	19.0
4.2099	5.0	940	5.8284	0.7498	0.0	0.7731	0.7387	15.8805
1.7867	6.0	1128	2.0588	0.7498	0.0	0.7731	0.7387	15.761
3.6568	7.0	1316	1.7242	0.7498	0.0	0.7731	0.7387	12.9371
4.1534	8.0	1504	1.4302	0.7498	0.0	0.7731	0.7387	15.1761
0.9298	9.0	1692	1.3871	0.7498	0.0	0.7731	0.7387	14.3711
1.5429	10.0	1880	1.1574	0.7498	0.0	0.7731	0.7387	18.4969
1.1307	11.0	2068	1.0076	0.8804	0.1144	0.8573	0.8336	19.0
0.9566	12.0	2256	0.9130	0.7498	0.0	0.7731	0.7387	19.0
0.6761	13.0	2444	0.8454	0.8804	0.1144	0.8573	0.8336	19.0
0.1983	14.0	2632	0.7835	0.0	0.0	0.0	0.0	19.0
0.5771	15.0	2820	0.7411	0.0	0.0	0.0	0.0	19.0
1.1574	16.0	3008	0.7099	0.3594	0.2648	0.3594	0.3594	19.0
0.1504	17.0	3196	0.6689	0.0	0.0	0.0	0.0	19.0
0.1265	18.0	3384	0.6576	0.3594	0.2648	0.3594	0.3594	19.0
0.6688	19.0	3572	0.6355	0.3594	0.2648	0.3594	0.3594	19.0
0.1103	20.0	3760	0.6101	0.5391	0.3972	0.5391	0.5391	19.0
0.4472	21.0	3948	0.6018	0.0	0.0	0.0	0.0	19.0
0.0696	22.0	4136	0.5804	0.2459	0.1324	0.2459	0.2223	19.0
0.2857	23.0	4324	0.5685	0.0	0.0	0.0	0.0	19.0
0.4744	24.0	4512	0.5475	0.2459	0.1324	0.2459	0.2223	19.0
0.2841	25.0	4700	0.5590	0.0	0.0	0.0	0.0	19.0

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0.dev20230811+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2