metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- billsum
metrics:
- rouge
model-index:
- name: my_awesome_billsum_model_custom_key
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: billsum
type: billsum
config: default
split: ca_test
args: default
metrics:
- name: Rouge1
type: rouge
value: 0.1384
my_awesome_billsum_model_custom_key
This model is a fine-tuned version of t5-small on the billsum dataset. It achieves the following results on the evaluation set:
- Loss: 2.5689
- Rouge1: 0.1384
- Rouge2: 0.0474
- Rougel: 0.1132
- Rougelsum: 0.1134
- 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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 62 | 2.8725 | 0.138 | 0.0449 | 0.1127 | 0.1127 | 19.0 |
| No log | 2.0 | 124 | 2.6478 | 0.1276 | 0.0387 | 0.105 | 0.1052 | 19.0 |
| No log | 3.0 | 186 | 2.5846 | 0.1343 | 0.0429 | 0.1088 | 0.109 | 19.0 |
| No log | 4.0 | 248 | 2.5689 | 0.1384 | 0.0474 | 0.1132 | 0.1134 | 19.0 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2