metadata

license: apache-2.0
tags:
  - summarization
  - generated_from_trainer
datasets:
  - amazon_reviews_multi
metrics:
  - rouge
model-index:
  - name: bart-base-finetuned-amazon-onlyen
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: amazon_reviews_multi
          type: amazon_reviews_multi
          args: en
        metrics:
          - name: Rouge1
            type: rouge
            value: 17.2662

bart-base-finetuned-amazon-onlyen

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

Loss: 3.7572
Rouge1: 17.2662
Rouge2: 8.7425
Rougel: 16.5765
Rougelsum: 16.6844

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: 5.6e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
2.9212	1.0	771	2.8034	15.381	8.5254	15.223	15.059
2.3109	2.0	1542	2.8386	19.8947	11.0965	19.4876	19.5366
1.8973	3.0	2313	2.9258	17.7443	8.9232	17.311	17.1796
1.5421	4.0	3084	3.0696	17.8204	8.8919	17.3889	17.205
1.2391	5.0	3855	3.2609	15.9828	8.0523	15.393	15.3808
0.9736	6.0	4626	3.4080	15.7572	8.806	15.2435	15.3036
0.7824	7.0	5397	3.5537	18.4389	9.5135	17.7836	17.8758
0.6233	8.0	6168	3.6909	14.6698	6.9584	13.9417	14.0057
0.5086	9.0	6939	3.7357	16.9465	7.7604	16.1993	16.2963
0.4412	10.0	7710	3.7572	17.2662	8.7425	16.5765	16.6844

Framework versions

Transformers 4.19.2
Pytorch 1.11.0+cu113
Datasets 2.2.2
Tokenizers 0.12.1