autoevaluate/summarization

1---2license: apache-2.03tags:4- generated_from_trainer5- summarization6datasets:7- xsum8- autoevaluate/xsum-sample9metrics:10- rouge11model-index:12- name: summarization13  results:14  - task:15      name: Sequence-to-sequence Language Modeling16      type: text2text-generation17    dataset:18      name: xsum19      type: xsum20      args: default21    metrics:22    - name: Rouge123      type: rouge24      value: 23.940525---26 27<!-- This model card has been generated automatically according to the information the Trainer had access to. You28should probably proofread and complete it, then remove this comment. -->29 30# summarization31 32This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the xsum dataset.33It achieves the following results on the evaluation set:34- Loss: 2.669035- Rouge1: 23.940536- Rouge2: 5.087937- Rougel: 18.498138- Rougelsum: 18.503239- Gen Len: 18.737640 41## Model description42 43More information needed44 45## Intended uses & limitations46 47More information needed48 49## Training and evaluation data50 51More information needed52 53## Training procedure54 55### Training hyperparameters56 57The following hyperparameters were used during training:58- learning_rate: 2e-0559- train_batch_size: 1660- eval_batch_size: 1661- seed: 4262- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-0863- lr_scheduler_type: linear64- training_steps: 100065- mixed_precision_training: Native AMP66 67### Training results68 69| Training Loss | Epoch | Step | Validation Loss | Rouge1  | Rouge2 | Rougel  | Rougelsum | Gen Len |70|:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:|71| 2.9249        | 0.08  | 1000 | 2.6690          | 23.9405 | 5.0879 | 18.4981 | 18.5032   | 18.7376 |72 73 74### Framework versions75 76- Transformers 4.19.277- Pytorch 1.11.0+cu11378- Datasets 2.2.279- Tokenizers 0.12.180