metadata
license: apache-2.0
base_model: t5-small
tags:
- generated_from_trainer
datasets:
- cnn_dailymail
metrics:
- rouge
model-index:
- name: T5_Model
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: cnn_dailymail
type: cnn_dailymail
config: 3.0.0
split: test
args: 3.0.0
metrics:
- name: Rouge1
type: rouge
value: 0.2473
T5_Model
This model is a fine-tuned version of t5-small on the cnn_dailymail dataset. It achieves the following results on the evaluation set:
- Loss: 1.7795
- Rouge1: 0.2473
- Rouge2: 0.1174
- Rougel: 0.2041
- Rougelsum: 0.2042
- Gen Len: 18.9999
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: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: tpu
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 2.0058 | 1.0 | 35890 | 1.8209 | 0.247 | 0.1174 | 0.2039 | 0.2039 | 18.9992 |
| 1.9949 | 2.0 | 71780 | 1.8004 | 0.2469 | 0.117 | 0.2036 | 0.2036 | 18.9995 |
| 1.948 | 3.0 | 107670 | 1.7938 | 0.2477 | 0.1176 | 0.2047 | 0.2047 | 18.9999 |
| 1.9459 | 4.0 | 143560 | 1.7884 | 0.2478 | 0.1182 | 0.2049 | 0.2049 | 18.9999 |
| 1.924 | 5.0 | 179450 | 1.7844 | 0.2477 | 0.1179 | 0.2045 | 0.2046 | 18.9996 |
| 1.9301 | 6.0 | 215340 | 1.7824 | 0.2477 | 0.1179 | 0.2044 | 0.2044 | 18.9999 |
| 1.9284 | 7.0 | 251230 | 1.7808 | 0.2474 | 0.1177 | 0.2044 | 0.2045 | 18.9999 |
| 1.9217 | 8.0 | 287120 | 1.7795 | 0.2473 | 0.1174 | 0.2041 | 0.2042 | 18.9999 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.0.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0