Hugging Face's logo

Lancelot53
/
flan-t5-base-xlsum

Text2Text Generation
Transformers
PyTorch
t5
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Community
1
flan-t5-base-xlsum / README.md
Lancelot53's picture
Lancelot53
update model card README.md
7538355
preview code
|
raw
history blame
6.91 kB
metadata
license: apache-2.0
base_model: google/flan-t5-base
tags:
  - generated_from_trainer
datasets:
  - xlsum
model-index:
  - name: flan-t5-base-xlsum
    results: []

flan-t5-base-xlsum

This model is a fine-tuned version of google/flan-t5-base on the xlsum dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3989

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: 5e-05
  • train_batch_size: 6
  • eval_batch_size: 12
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 200
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss
29.1967 0.05 100 4.4538
2.3457 0.09 200 0.4789
0.5064 0.14 300 0.4116
0.4857 0.18 400 0.4077
0.4749 0.23 500 0.4069
0.4695 0.28 600 0.4059
0.4754 0.32 700 0.4041
0.4609 0.37 800 0.4044
0.4703 0.42 900 0.4037
0.4656 0.46 1000 0.4025
0.464 0.51 1100 0.4020
0.4679 0.55 1200 0.4026
0.462 0.6 1300 0.4021
0.4665 0.65 1400 0.4006
0.4617 0.69 1500 0.4013
0.4541 0.74 1600 0.4000
0.4566 0.79 1700 0.3997
0.4646 0.83 1800 0.3995
0.4445 0.88 1900 0.3999
0.4733 0.92 2000 0.3993
0.4703 0.97 2100 0.3995
0.4412 1.02 2200 0.3998
0.4249 1.06 2300 0.4000
0.436 1.11 2400 0.3995
0.4333 1.16 2500 0.3989
0.4249 1.2 2600 0.3984
0.4312 1.25 2700 0.3988
0.4376 1.29 2800 0.3992
0.4276 1.34 2900 0.3990
0.4258 1.39 3000 0.3983
0.4411 1.43 3100 0.3986
0.4352 1.48 3200 0.3989
0.4429 1.53 3300 0.3974
0.4466 1.57 3400 0.3980
0.4311 1.62 3500 0.3977
0.427 1.66 3600 0.3976
0.4433 1.71 3700 0.3977
0.4228 1.76 3800 0.3984
0.4247 1.8 3900 0.3980
0.4275 1.85 4000 0.3980
0.4523 1.9 4100 0.3970
0.4258 1.94 4200 0.3976
0.4329 1.99 4300 0.3978
0.4146 2.03 4400 0.3988
0.4025 2.08 4500 0.3997
0.3944 2.13 4600 0.3989
0.4034 2.17 4700 0.3984
0.4099 2.22 4800 0.3987
0.3989 2.27 4900 0.3983
0.4269 2.31 5000 0.3990
0.4273 2.36 5100 0.3988
0.4117 2.4 5200 0.3981
0.4117 2.45 5300 0.3984
0.4037 2.5 5400 0.3978
0.4158 2.54 5500 0.3981
0.4081 2.59 5600 0.3982
0.4125 2.64 5700 0.3982
0.4086 2.68 5800 0.3988
0.4143 2.73 5900 0.3986
0.4025 2.77 6000 0.3981
0.4141 2.82 6100 0.3979
0.4239 2.87 6200 0.3975
0.4217 2.91 6300 0.3979
0.4099 2.96 6400 0.3972
0.4008 3.01 6500 0.3977
0.4092 3.05 6600 0.3998
0.3898 3.1 6700 0.4000
0.3978 3.14 6800 0.3985
0.4004 3.19 6900 0.3996
0.3998 3.24 7000 0.3996
0.3908 3.28 7100 0.3993
0.4021 3.33 7200 0.3994
0.3889 3.37 7300 0.3993
0.4009 3.42 7400 0.3984
0.3835 3.47 7500 0.3988
0.3999 3.51 7600 0.3986
0.409 3.56 7700 0.3985
0.3927 3.61 7800 0.3984
0.407 3.65 7900 0.3980
0.389 3.7 8000 0.3989
0.3976 3.74 8100 0.3981
0.4075 3.79 8200 0.3982
0.3897 3.84 8300 0.3981
0.3805 3.88 8400 0.3983
0.393 3.93 8500 0.3983
0.398 3.98 8600 0.3980
0.3832 4.02 8700 0.3985
0.384 4.07 8800 0.3989
0.3787 4.11 8900 0.3989
0.3816 4.16 9000 0.3994
0.3857 4.21 9100 0.3991
0.3909 4.25 9200 0.3990
0.3858 4.3 9300 0.3993
0.4021 4.35 9400 0.3993
0.3879 4.39 9500 0.3991
0.3752 4.44 9600 0.3994
0.3882 4.48 9700 0.3994
0.3881 4.53 9800 0.3992
0.4089 4.58 9900 0.3988
0.3801 4.62 10000 0.3989
0.3925 4.67 10100 0.3989
0.3858 4.72 10200 0.3990
0.3883 4.76 10300 0.3988
0.3808 4.81 10400 0.3989
0.4012 4.85 10500 0.3989
0.384 4.9 10600 0.3989
0.3828 4.95 10700 0.3989
0.3899 4.99 10800 0.3989

Framework versions

  • Transformers 4.31.0
  • Pytorch 2.0.1+cu117
  • Datasets 2.13.1
  • Tokenizers 0.13.3