README.md

metadata

base_model: weny22/sum_model_t5_saved
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: extract_long_text_balanced_data
    results: []

extract_long_text_balanced_data

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: 2.3331
• Rouge1: 0.2094
• Rouge2: 0.0794
• Rougel: 0.1697
• Rougelsum: 0.1696
• Gen Len: 18.9853

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	119	2.3318	0.18	0.0552	0.1439	0.1439	18.97
No log	2.0	238	2.1980	0.1899	0.06	0.1503	0.1503	18.972
No log	3.0	357	2.1448	0.1952	0.0646	0.1542	0.1541	18.9993
No log	4.0	476	2.1372	0.1983	0.0683	0.1574	0.1574	18.9453
2.8015	5.0	595	2.1142	0.2	0.0725	0.1611	0.1611	18.9933
2.8015	6.0	714	2.0970	0.2027	0.0757	0.1629	0.1629	18.9987
2.8015	7.0	833	2.1187	0.2027	0.0755	0.1637	0.1634	18.968
2.8015	8.0	952	2.1222	0.2013	0.0737	0.1619	0.1618	18.9753
2.02	9.0	1071	2.1316	0.2021	0.0764	0.1648	0.1647	18.9667
2.02	10.0	1190	2.1455	0.2109	0.0784	0.169	0.1689	18.982
2.02	11.0	1309	2.1580	0.2065	0.0781	0.167	0.1669	18.968
2.02	12.0	1428	2.1792	0.2088	0.0788	0.1693	0.169	18.9767
1.683	13.0	1547	2.1958	0.2085	0.0781	0.1689	0.1689	18.9913
1.683	14.0	1666	2.2436	0.2082	0.0785	0.1693	0.1692	18.978
1.683	15.0	1785	2.2480	0.2075	0.0797	0.1678	0.1678	18.9853
1.683	16.0	1904	2.2714	0.208	0.0789	0.1686	0.1685	18.9887
1.45	17.0	2023	2.2771	0.2091	0.0787	0.1693	0.1691	18.98
1.45	18.0	2142	2.2913	0.2103	0.0792	0.17	0.1698	18.9873
1.45	19.0	2261	2.3163	0.2094	0.0792	0.1699	0.1697	18.9893
1.45	20.0	2380	2.3331	0.2094	0.0794	0.1697	0.1696	18.9853

Framework versions

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