README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - billsum
metrics:
  - rouge
model-index:
  - name: my_awesome_eli5_clm-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: 0.1959

my_awesome_eli5_clm-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: 2.3071
• Rouge1: 0.1959
• Rouge2: 0.1013
• Rougel: 0.1685
• Rougelsum: 0.1683
• 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	62	2.7637	0.1277	0.0387	0.1065	0.1066	19.0
No log	2.0	124	2.5350	0.1408	0.0506	0.1165	0.1165	19.0
No log	3.0	186	2.4431	0.1503	0.0589	0.1245	0.1245	19.0
No log	4.0	248	2.3946	0.1774	0.0796	0.1502	0.1501	19.0
No log	5.0	310	2.3601	0.19	0.0939	0.1631	0.1631	19.0
No log	6.0	372	2.3400	0.1952	0.0993	0.1676	0.1676	19.0
No log	7.0	434	2.3238	0.196	0.1003	0.1682	0.1681	19.0
No log	8.0	496	2.3140	0.1973	0.1017	0.1693	0.1692	19.0
2.7599	9.0	558	2.3084	0.1957	0.1009	0.1686	0.1682	19.0
2.7599	10.0	620	2.3071	0.1959	0.1013	0.1685	0.1683	19.0

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3