README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - billsum
metrics:
  - rouge
model-index:
  - name: custom_billsum_model
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: billsum
          type: billsum
          config: default
          split: ca_test
          args: default
        metrics:
          - name: Rouge1
            type: rouge
            value: 0.1968

custom_billsum_model

This model is a fine-tuned version of t5-small on the billsum dataset. It achieves the following results on the evaluation set:

• Loss: 2.2150
• Rouge1: 0.1968
• Rouge2: 0.0981
• Rougel: 0.1672
• Rougelsum: 0.167
• 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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• 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	62	2.4551	0.1626	0.0663	0.135	0.135	19.0
No log	2.0	124	2.3987	0.1882	0.0866	0.1577	0.1577	19.0
No log	3.0	186	2.3639	0.1964	0.0937	0.1652	0.165	19.0
No log	4.0	248	2.3370	0.1943	0.0931	0.164	0.1638	19.0
No log	5.0	310	2.3135	0.1942	0.0938	0.1646	0.1643	19.0
No log	6.0	372	2.2949	0.195	0.0938	0.1648	0.1648	19.0
No log	7.0	434	2.2809	0.1937	0.0944	0.1643	0.1642	19.0
No log	8.0	496	2.2676	0.1949	0.0957	0.1664	0.166	19.0
2.5047	9.0	558	2.2582	0.1954	0.097	0.1665	0.1662	19.0
2.5047	10.0	620	2.2510	0.1951	0.0966	0.1661	0.166	19.0
2.5047	11.0	682	2.2416	0.1962	0.0979	0.1673	0.1671	19.0
2.5047	12.0	744	2.2360	0.196	0.0975	0.1664	0.1663	19.0
2.5047	13.0	806	2.2302	0.1965	0.098	0.1667	0.1666	19.0
2.5047	14.0	868	2.2262	0.1973	0.0985	0.1673	0.1671	19.0
2.5047	15.0	930	2.2224	0.197	0.0976	0.1668	0.1667	19.0
2.5047	16.0	992	2.2200	0.1973	0.0984	0.1673	0.1671	19.0
2.3391	17.0	1054	2.2183	0.1967	0.0974	0.1669	0.1666	19.0
2.3391	18.0	1116	2.2164	0.1968	0.0974	0.1669	0.1666	19.0
2.3391	19.0	1178	2.2152	0.1969	0.0982	0.1673	0.1671	19.0
2.3391	20.0	1240	2.2150	0.1968	0.0981	0.1672	0.167	19.0

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3