mt5-base-finetuned-test_30483_prefix_summarize-finetuned-test_21911_prefix_summarize_V12159

This model is a fine-tuned version of emilstabil/mt5-base-finetuned-test_30483_prefix_summarize-finetuned-test_21911_prefix_summarize on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.6592
• Rouge1: 28.7045
• Rouge2: 7.6597
• Rougel: 17.7911
• Rougelsum: 26.2935
• Gen Len: 89.74

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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.8662	1.25	500	2.3721	27.5477	7.8163	17.0749	25.3524	90.59
1.826	2.5	1000	2.3866	27.1584	7.2417	16.6686	24.744	88.96
1.7636	3.75	1500	2.3649	27.1333	8.0536	17.3248	24.7674	84.37
1.7248	5.0	2000	2.3946	26.1796	7.2833	16.7034	23.9083	82.43
1.6394	6.25	2500	2.4257	27.996	7.9177	17.3958	25.8586	89.47
1.6023	7.5	3000	2.4280	27.6978	7.8114	17.6006	25.6393	89.44
1.5606	8.75	3500	2.4460	28.2604	7.7071	17.2627	26.0593	98.12
1.5173	10.0	4000	2.4752	26.6534	7.5213	17.2009	24.433	80.49
1.4655	11.25	4500	2.4787	28.6229	8.5816	17.6534	26.2693	90.87
1.4405	12.5	5000	2.4944	28.8101	8.5897	17.964	26.6235	89.81
1.4015	13.75	5500	2.5129	27.4546	7.67	17.5582	25.2634	86.03
1.3729	15.0	6000	2.5276	28.7309	8.2089	18.2854	26.5967	91.97
1.3293	16.25	6500	2.5434	27.4011	7.4927	17.5076	25.068	84.97
1.3279	17.5	7000	2.5583	28.4385	7.6384	17.5373	26.0499	92.71
1.2727	18.75	7500	2.5929	28.9833	7.9095	17.4616	26.7396	94.69
1.2693	20.0	8000	2.5798	27.9046	7.7553	17.2599	25.705	85.16
1.2416	21.25	8500	2.6174	27.7	7.8456	17.4919	25.6451	84.51
1.2174	22.5	9000	2.6227	28.5114	7.6761	17.4532	26.1349	91.37
1.2088	23.75	9500	2.6222	28.2178	7.5203	17.5631	25.8688	87.71
1.1982	25.0	10000	2.6334	28.9268	7.7591	17.8701	26.5417	94.73
1.1839	26.25	10500	2.6403	29.1922	7.691	17.8536	26.9276	92.78
1.1589	27.5	11000	2.6541	28.9328	7.9311	17.8917	26.81	92.8
1.1633	28.75	11500	2.6658	28.7038	7.618	17.9869	26.4802	94.12
1.1605	30.0	12000	2.6592	28.7045	7.6597	17.7911	26.2935	89.74

Framework versions

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