Hugging Face's logo

jhpassion0621
/
kp-umt5-base

Text2Text Generation
Transformers
TensorBoard
Safetensors
umt5
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
kp-umt5-base / README.md
jhpassion0621's picture
jhpassion0621
End of training
a2bc661 verified
preview code
|
raw
history blame
3.87 kB
metadata
license: apache-2.0
base_model: google/umt5-base
tags:
  - generated_from_trainer
metrics:
  - bleu
model-index:
  - name: kp-umt5-base
    results: []

kp-umt5-base

This model is a fine-tuned version of google/umt5-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.5488
  • Bleu: 9.2954
  • Gen Len: 18.0013

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: 1

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
4.1202 0.03 1000 2.9777 2.8238 18.173
3.5142 0.05 2000 2.5373 4.457 18.0813
3.1913 0.08 3000 2.3418 5.029 18.019
3.0343 0.11 4000 2.2078 5.5467 18.163
2.9 0.14 5000 2.1214 5.9178 18.14
2.7425 0.16 6000 2.0479 6.1516 18.0753
2.6798 0.19 7000 1.9939 6.4309 18.1157
2.6029 0.22 8000 1.9368 6.7683 18.1533
2.5378 0.24 9000 1.8959 6.9762 18.1603
2.5132 0.27 10000 1.8602 7.3219 18.2093
2.4394 0.3 11000 1.8264 7.3572 18.158
2.4158 0.33 12000 1.7959 7.5914 18.1257
2.3339 0.35 13000 1.7696 7.6829 18.077
2.3397 0.38 14000 1.7440 7.87 18.0993
2.293 0.41 15000 1.7219 8.1915 18.1303
2.2591 0.44 16000 1.7057 8.2801 18.0963
2.274 0.46 17000 1.6874 8.4263 18.0953
2.2387 0.49 18000 1.6709 8.5568 18.0837
2.2176 0.52 19000 1.6527 8.6313 18.0767
2.1742 0.54 20000 1.6414 8.718 18.0613
2.2095 0.57 21000 1.6260 8.8699 18.044
2.1936 0.6 22000 1.6154 8.8912 18.0573
2.0923 0.63 23000 1.6119 8.9109 18.091
2.1835 0.65 24000 1.6015 8.9474 18.0533
2.1374 0.68 25000 1.5962 9.0021 18.064
2.1286 0.71 26000 1.5816 9.0722 18.06
2.0589 0.73 27000 1.5787 9.1697 18.0667
2.0535 0.76 28000 1.5755 9.2336 18.05
2.1078 0.79 29000 1.5697 9.1774 18.0537
2.05 0.82 30000 1.5635 9.2354 18.0347
2.0517 0.84 31000 1.5593 9.2574 18.0013
2.0802 0.87 32000 1.5549 9.2727 18.0113
2.0137 0.9 33000 1.5540 9.2265 18.0007
2.069 0.92 34000 1.5497 9.3092 18.019
2.0969 0.95 35000 1.5498 9.2936 18.0113
2.0911 0.98 36000 1.5488 9.2954 18.0013

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.1