README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - aeslc
metrics:
  - rouge
model-index:
  - name: bart-large-finetuned-aeslc
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: aeslc
          type: aeslc
          config: default
          split: validation
          args: default
        metrics:
          - name: Rouge1
            type: rouge
            value: 38.0679

bart-large-finetuned-aeslc

This model is a fine-tuned version of facebook/bart-large on the aeslc dataset. It achieves the following results on the evaluation set:

• Loss: 4.6657
• Rouge1: 38.0679
• Rouge2: 19.8904
• Rougel: 37.1179
• Rougelsum: 37.1066
• Gen Len: 9.1316

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: 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: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.6268	1.0	7218	4.4738	36.0615	18.1082	35.2023	35.175	8.0694
1.9548	2.0	14436	4.4965	37.1784	19.2713	36.2803	36.2347	9.2699
1.4117	3.0	21654	4.6657	38.0679	19.8904	37.1179	37.1066	9.1316

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3