Edit model card

t5-summ

This model is a fine-tuned version of t5-small on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.0871
  • Rouge1: 0.1961
  • Rouge2: 0.099
  • Rougel: 0.1691
  • Rougelsum: 0.1691
  • 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: 30
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
No log 1.0 62 2.6868 0.1256 0.0388 0.1039 0.104 19.0
No log 2.0 124 2.4594 0.143 0.0547 0.1191 0.1193 19.0
No log 3.0 186 2.3653 0.1677 0.0718 0.1396 0.1396 19.0
No log 4.0 248 2.3113 0.1913 0.0917 0.1613 0.161 19.0
No log 5.0 310 2.2735 0.196 0.0974 0.1665 0.1663 19.0
No log 6.0 372 2.2417 0.1972 0.0996 0.1687 0.1686 19.0
No log 7.0 434 2.2197 0.1985 0.1011 0.17 0.1699 19.0
No log 8.0 496 2.2011 0.1982 0.1012 0.1698 0.1697 19.0
2.7383 9.0 558 2.1829 0.198 0.1 0.1698 0.1698 19.0
2.7383 10.0 620 2.1724 0.1985 0.1011 0.1703 0.1702 19.0
2.7383 11.0 682 2.1605 0.1991 0.1017 0.1708 0.1709 19.0
2.7383 12.0 744 2.1489 0.1992 0.1022 0.1717 0.1719 19.0
2.7383 13.0 806 2.1420 0.1994 0.1028 0.1716 0.1716 19.0
2.7383 14.0 868 2.1322 0.2003 0.1041 0.1726 0.1726 19.0
2.7383 15.0 930 2.1265 0.2 0.103 0.172 0.1719 19.0
2.7383 16.0 992 2.1196 0.1993 0.1014 0.1718 0.1718 19.0
2.3748 17.0 1054 2.1165 0.1979 0.1011 0.1709 0.1709 19.0
2.3748 18.0 1116 2.1090 0.1985 0.1011 0.1701 0.1703 19.0
2.3748 19.0 1178 2.1063 0.1984 0.1014 0.1706 0.1708 19.0
2.3748 20.0 1240 2.1031 0.1993 0.1031 0.1714 0.1715 19.0
2.3748 21.0 1302 2.0997 0.1982 0.1018 0.1707 0.1708 19.0
2.3748 22.0 1364 2.0970 0.1966 0.1002 0.1692 0.1694 19.0
2.3748 23.0 1426 2.0951 0.1948 0.0986 0.1681 0.1682 19.0
2.3748 24.0 1488 2.0928 0.1959 0.0995 0.1691 0.1693 19.0
2.2969 25.0 1550 2.0919 0.1958 0.0995 0.1689 0.169 19.0
2.2969 26.0 1612 2.0892 0.1955 0.099 0.1687 0.1688 19.0
2.2969 27.0 1674 2.0883 0.196 0.0994 0.1692 0.1692 19.0
2.2969 28.0 1736 2.0877 0.1959 0.0994 0.1692 0.1693 19.0
2.2969 29.0 1798 2.0871 0.196 0.0995 0.1692 0.1692 19.0
2.2969 30.0 1860 2.0871 0.1961 0.099 0.1691 0.1691 19.0

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.1.2
  • Datasets 2.18.0
  • Tokenizers 0.15.2
Downloads last month
4
Safetensors
Model size
60.5M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for suneeln-duke/t5-summ

Base model

google-t5/t5-small
Finetuned
(1531)
this model