sum_model_2r1e_3_20_extract_long_text

only extract text > 500 word

number of sentence in train data = 12 number of sentence in validation data = 12 words in train data: 350 words in summury: 32 after modification: 294 words in validation data: 344 words in summury: 32 after modification 294

This model is a fine-tuned version of weny22/sum_model_t5_saved on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.0769
• Rouge1: 0.2174
• Rouge2: 0.0883
• Rougel: 0.1783
• Rougelsum: 0.1783
• Gen Len: 18.9707

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: 0.002
• train_batch_size: 64
• eval_batch_size: 64
• 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
No log	1.0	335	2.1718	0.1974	0.0646	0.1569	0.1572	18.9793
2.7704	2.0	670	2.0191	0.2026	0.0737	0.1627	0.1629	18.99
2.1753	3.0	1005	1.9912	0.209	0.0792	0.1703	0.1704	18.9607
2.1753	4.0	1340	1.9403	0.2051	0.0797	0.1673	0.1673	18.9793
1.9544	5.0	1675	1.9172	0.2162	0.0857	0.1766	0.1766	18.9853
1.8287	6.0	2010	1.9292	0.2115	0.0829	0.1719	0.172	18.9727
1.8287	7.0	2345	1.9303	0.2113	0.0838	0.1722	0.1721	18.972
1.6903	8.0	2680	1.9165	0.2146	0.0843	0.1754	0.1756	18.972
1.6063	9.0	3015	1.9242	0.2166	0.0864	0.1764	0.1765	18.9707
1.6063	10.0	3350	1.9196	0.2143	0.0843	0.175	0.1749	18.986
1.4965	11.0	3685	1.9313	0.2148	0.0847	0.1743	0.1742	18.9893
1.4333	12.0	4020	1.9512	0.2165	0.0873	0.1777	0.1778	18.99
1.4333	13.0	4355	1.9756	0.2163	0.0882	0.1779	0.178	18.9747
1.346	14.0	4690	1.9945	0.2182	0.0893	0.1792	0.1792	18.9873
1.2887	15.0	5025	2.0073	0.2154	0.085	0.1756	0.1758	18.9873
1.2887	16.0	5360	2.0176	0.2166	0.0878	0.1772	0.1773	18.9687
1.2234	17.0	5695	2.0302	0.2172	0.0873	0.178	0.178	18.9767
1.1833	18.0	6030	2.0505	0.2156	0.0854	0.1761	0.1761	18.9653
1.1833	19.0	6365	2.0633	0.2174	0.0875	0.1777	0.1778	18.9787
1.1396	20.0	6700	2.0769	0.2174	0.0883	0.1783	0.1783	18.9707

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2