metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- govreport-summarization
metrics:
- rouge
model-index:
- name: flan-t5-gov-report-sum
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: govreport-summarization
type: govreport-summarization
config: document
split: test
args: document
metrics:
- name: Rouge1
type: rouge
value: 5.8729
flan-t5-gov-report-sum
This model is a fine-tuned version of google/flan-t5-base on the govreport-summarization dataset. It achieves the following results on the evaluation set:
- Loss: 2.2385
- Rouge1: 5.8729
- Rouge2: 3.0763
- Rougel: 5.1016
- Rougelsum: 5.646
- 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: 5e-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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 2.5801 | 1.0 | 2190 | 2.3211 | 5.6226 | 2.9142 | 4.9535 | 5.417 | 19.0 |
| 2.5125 | 2.0 | 4380 | 2.2748 | 5.7982 | 3.0365 | 5.0726 | 5.5837 | 19.0 |
| 2.453 | 3.0 | 6570 | 2.2545 | 5.8744 | 3.0997 | 5.1196 | 5.6524 | 19.0 |
| 2.436 | 4.0 | 8760 | 2.2430 | 5.8669 | 3.0525 | 5.0849 | 5.631 | 19.0 |
| 2.4144 | 5.0 | 10950 | 2.2385 | 5.8729 | 3.0763 | 5.1016 | 5.646 | 19.0 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu102
- Datasets 2.9.0
- Tokenizers 0.13.2