mt5-base-finetuned-test_30483_prefix_summarize-finetuned-test_21911_prefix_summarize

This model is a fine-tuned version of emilstabil/mt5-base-finetuned-test_30483_prefix_summarize on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.3442
• Rouge1: 26.765
• Rouge2: 7.4581
• Rougel: 17.0907
• Rougelsum: 24.6176
• Gen Len: 86.94

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: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.5963	1.25	500	2.3638	24.4676	7.3477	16.2948	22.5499	75.59
2.4234	2.5	1000	2.3433	25.3322	7.2863	16.2352	23.3304	76.8
2.3196	3.75	1500	2.3308	25.5632	7.1264	16.4164	23.4946	82.05
2.2519	5.0	2000	2.3315	25.0637	7.1123	16.1223	23.0408	74.67
2.1623	6.25	2500	2.3487	26.4375	7.3373	16.6264	24.2391	80.73
2.1145	7.5	3000	2.3376	25.703	7.2824	16.5698	23.5391	77.68
2.0668	8.75	3500	2.3416	25.5062	7.2385	16.4797	23.3875	77.94
2.0175	10.0	4000	2.3330	26.1031	7.581	16.7435	23.8668	79.52
1.9757	11.25	4500	2.3453	26.8941	7.3977	16.6922	24.4811	89.72
1.9527	12.5	5000	2.3354	27.0584	7.7057	17.3043	24.8764	83.53
1.9263	13.75	5500	2.3444	26.7263	7.5763	16.9772	24.4145	85.38
1.9072	15.0	6000	2.3393	26.6875	7.3326	16.9405	24.4442	87.55
1.8678	16.25	6500	2.3390	26.1098	7.2367	16.6426	24.0273	83.85
1.882	17.5	7000	2.3388	26.6687	7.543	17.1994	24.5235	84.56
1.8269	18.75	7500	2.3447	27.3232	7.8632	17.3518	25.1263	87.26
1.8403	20.0	8000	2.3442	26.765	7.4581	17.0907	24.6176	86.94

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0
• Datasets 2.12.0
• Tokenizers 0.13.3