Hugging Face's logo

arthurmluz
/
ptt5-xlsumm-gptextsum

Text2Text Generation
Transformers
PyTorch
t5
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Community
ptt5-xlsumm-gptextsum / README.md
arthurmluz's picture
arthurmluz
Model save
0790926
preview code
|
raw
history blame
4.48 kB
metadata
license: mit
base_model: arthurmluz/ptt5-xlsumm-30epochs
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: ptt5-xlsumm-gptextsum
    results: []

ptt5-xlsumm-gptextsum

This model is a fine-tuned version of arthurmluz/ptt5-xlsumm-30epochs on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.1132
  • Rouge1: 0.1715
  • Rouge2: 0.0919
  • Rougel: 0.1417
  • Rougelsum: 0.1611
  • 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: 2
  • eval_batch_size: 2
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 30

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
No log 1.0 70 2.3037 0.1617 0.0691 0.1287 0.1467 19.0
No log 2.0 140 2.2106 0.1722 0.082 0.1362 0.1585 19.0
2.3539 3.0 210 2.1604 0.1738 0.0854 0.1387 0.1604 19.0
2.3539 4.0 280 2.1325 0.1727 0.0868 0.1407 0.1632 19.0
2.3539 5.0 350 2.1117 0.1707 0.0875 0.137 0.1603 19.0
2.0032 6.0 420 2.0957 0.1723 0.0907 0.1415 0.1607 19.0
2.0032 7.0 490 2.0848 0.1722 0.09 0.1414 0.1617 19.0
2.0032 8.0 560 2.0790 0.1757 0.0918 0.1429 0.164 19.0
1.8158 9.0 630 2.0800 0.1723 0.0929 0.1421 0.1614 19.0
1.8158 10.0 700 2.0736 0.1733 0.0923 0.1428 0.1617 19.0
1.8158 11.0 770 2.0721 0.1755 0.0955 0.1454 0.1631 19.0
1.6764 12.0 840 2.0784 0.1763 0.0973 0.1459 0.1637 19.0
1.6764 13.0 910 2.0761 0.1752 0.094 0.1456 0.1638 19.0
1.6764 14.0 980 2.0802 0.1745 0.0951 0.145 0.1631 19.0
1.5616 15.0 1050 2.0790 0.1745 0.0952 0.1458 0.1632 19.0
1.5616 16.0 1120 2.0841 0.1735 0.0946 0.1447 0.1629 19.0
1.5616 17.0 1190 2.0904 0.1731 0.0943 0.1444 0.1622 19.0
1.4821 18.0 1260 2.0909 0.1727 0.0934 0.1433 0.1613 19.0
1.4821 19.0 1330 2.0934 0.1738 0.0948 0.1448 0.1632 19.0
1.4256 20.0 1400 2.0948 0.1726 0.0935 0.1434 0.1621 19.0
1.4256 21.0 1470 2.0981 0.173 0.0942 0.1435 0.1621 19.0
1.4256 22.0 1540 2.1023 0.1734 0.0945 0.1445 0.1631 19.0
1.3691 23.0 1610 2.1048 0.1726 0.0941 0.1436 0.1616 19.0
1.3691 24.0 1680 2.1058 0.1721 0.0948 0.1435 0.1619 19.0
1.3691 25.0 1750 2.1095 0.1721 0.0945 0.1435 0.1619 19.0
1.3444 26.0 1820 2.1103 0.1721 0.0948 0.1436 0.1624 19.0
1.3444 27.0 1890 2.1113 0.1715 0.0923 0.1417 0.1611 19.0
1.3444 28.0 1960 2.1121 0.1715 0.0919 0.1417 0.1611 19.0
1.3145 29.0 2030 2.1130 0.172 0.093 0.1425 0.1619 19.0
1.3145 30.0 2100 2.1132 0.1715 0.0919 0.1417 0.1611 19.0

Framework versions

  • Transformers 4.34.0
  • Pytorch 2.0.1+cu117
  • Datasets 2.14.5
  • Tokenizers 0.14.1