metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- billsum
metrics:
- rouge
- bleu
model-index:
- name: flan-t5-base-billsum_model
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: billsum
type: billsum
config: default
split: test
args: default
metrics:
- name: Rouge1
type: rouge
value: 0.2154
- name: Bleu
type: bleu
value: 0.0011
flan-t5-base-billsum_model
This model is a fine-tuned version of google/flan-t5-base on the billsum dataset. It achieves the following results on the evaluation set:
- Loss: nan
- Rouge1: 0.2154
- Rouge2: 0.1259
- Rougel: 0.1843
- Rougelsum: 0.1843
- Gen Len: 17.3735
- Bleu: 0.0011
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 40
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | Bleu |
|---|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 296 | nan | 0.2154 | 0.1259 | 0.1843 | 0.1843 | 17.3735 | 0.0011 |
| 0.0 | 2.0 | 592 | nan | 0.2154 | 0.1259 | 0.1843 | 0.1843 | 17.3735 | 0.0011 |
| 0.0 | 3.0 | 888 | nan | 0.2154 | 0.1259 | 0.1843 | 0.1843 | 17.3735 | 0.0011 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3