metadata

license: apache-2.0
base_model: google/long-t5-tglobal-base
tags:
  - summarization
  - generated_from_trainer
datasets:
  - gov_report_summarization_dataset
metrics:
  - rouge
model-index:
  - name: long-t5-tglobal-base-finetuned-govReport-4096
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: gov_report_summarization_dataset
          type: gov_report_summarization_dataset
          config: document
          split: validation
          args: document
        metrics:
          - name: Rouge1
            type: rouge
            value: 0.0463

long-t5-tglobal-base-finetuned-govReport-4096

This model is a fine-tuned version of google/long-t5-tglobal-base on the gov_report_summarization_dataset dataset. It achieves the following results on the evaluation set:

Loss: 1.3116
Rouge1: 0.0463
Rouge2: 0.0231
Rougel: 0.039
Rougelsum: 0.0436

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: 4e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 32
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
14.9489	1.0	25	2.8019	0.0416	0.0156	0.0343	0.0383
3.9751	2.0	50	2.0899	0.0396	0.0145	0.0332	0.0367
3.0361	3.0	75	1.4337	0.0395	0.0138	0.033	0.0366
1.8477	4.0	100	1.3654	0.0406	0.0157	0.0342	0.0378
1.6709	5.0	125	1.3355	0.0419	0.0171	0.036	0.0394
1.5673	6.0	150	1.3229	0.0446	0.022	0.0383	0.0423
1.5567	7.0	175	1.3181	0.047	0.0233	0.0393	0.0441
1.5066	8.0	200	1.3135	0.0467	0.0233	0.0394	0.0441
1.515	9.0	225	1.3117	0.0462	0.0231	0.0389	0.0436
1.4868	10.0	250	1.3116	0.0463	0.0231	0.039	0.0436

Framework versions

Transformers 4.37.0
Pytorch 2.1.2
Datasets 2.1.0
Tokenizers 0.15.1