metadata
license: apache-2.0
base_model: t5-small
tags:
- generated_from_trainer
datasets:
- billsum
metrics:
- rouge
model-index:
- name: my_awesome_billsum_model
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: 0.1993
my_awesome_billsum_model
This model is a fine-tuned version of t5-small on the billsum dataset. It achieves the following results on the evaluation set:
- Loss: 2.3574
- Rouge1: 0.1993
- Rouge2: 0.1009
- Rougel: 0.1702
- Rougelsum: 0.1704
- 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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
---|---|---|---|---|---|---|---|---|
No log | 1.0 | 248 | 2.5145 | 0.1394 | 0.0522 | 0.1171 | 0.1172 | 19.0 |
No log | 2.0 | 496 | 2.4057 | 0.1935 | 0.0951 | 0.1642 | 0.1644 | 19.0 |
2.8828 | 3.0 | 744 | 2.3667 | 0.2004 | 0.1024 | 0.1716 | 0.1717 | 19.0 |
2.8828 | 4.0 | 992 | 2.3574 | 0.1993 | 0.1009 | 0.1702 | 0.1704 | 19.0 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.14.0