metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- scitldr
metrics:
- rouge
model-index:
- name: paper-summary
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: scitldr
type: scitldr
config: Abstract
split: train
args: Abstract
metrics:
- name: Rouge1
type: rouge
value: 0.2967
paper-summary
This model is a fine-tuned version of t5-small on the scitldr dataset. It achieves the following results on the evaluation set:
- Loss: 3.1243
- Rouge1: 0.2967
- Rouge2: 0.1277
- Rougel: 0.2457
- Rougelsum: 0.2602
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: 5.6e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- 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 |
---|---|---|---|---|---|---|---|
2.9845 | 1.0 | 63 | 3.2358 | 0.288 | 0.1219 | 0.238 | 0.2522 |
2.7469 | 2.0 | 126 | 3.1815 | 0.2922 | 0.1251 | 0.2415 | 0.2562 |
2.7032 | 3.0 | 189 | 3.1564 | 0.2991 | 0.1303 | 0.2474 | 0.2619 |
2.6479 | 4.0 | 252 | 3.1354 | 0.2973 | 0.1282 | 0.2457 | 0.2608 |
2.6303 | 5.0 | 315 | 3.1297 | 0.2964 | 0.1272 | 0.2451 | 0.2598 |
2.6115 | 6.0 | 378 | 3.1243 | 0.2967 | 0.1277 | 0.2457 | 0.2602 |
2.5704 | 7.0 | 441 | 3.1264 | 0.2967 | 0.127 | 0.2453 | 0.26 |
2.6028 | 8.0 | 504 | 3.1252 | 0.2971 | 0.1275 | 0.2456 | 0.2604 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.1