Hugging Face's logo

roequitz
/
t5-abs-2209-2245-lr-0.001-bs-10-maxep-20

Safetensors
t5
Generated from Trainer
Model card Files Community
t5-abs-2209-2245-lr-0.001-bs-10-maxep-20 / README.md
roequitz's picture
roequitz
End of training
ae22b15 verified
preview code
|
raw
history blame contribute delete
6.41 kB
metadata
license: apache-2.0
base_model: google-t5/t5-base
tags:
  - generated_from_trainer
model-index:
  - name: t5-abs-2209-2245-lr-0.001-bs-10-maxep-20
    results: []

t5-abs-2209-2245-lr-0.001-bs-10-maxep-20

This model is a fine-tuned version of google-t5/t5-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 2.6565
  • Rouge/rouge1: 0.3034
  • Rouge/rouge2: 0.116
  • Rouge/rougel: 0.2608
  • Rouge/rougelsum: 0.2609
  • Bertscore/bertscore-precision: 0.789
  • Bertscore/bertscore-recall: 0.8697
  • Bertscore/bertscore-f1: 0.8251
  • Meteor: 0.2559
  • Gen Len: 60.4545

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: 0.001
  • train_batch_size: 10
  • eval_batch_size: 10
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 20
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 20
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Rouge/rouge1 Rouge/rouge2 Rouge/rougel Rouge/rougelsum Bertscore/bertscore-precision Bertscore/bertscore-recall Bertscore/bertscore-f1 Meteor Gen Len
1.5977 0.9885 43 1.8385 0.4519 0.2025 0.3799 0.3812 0.8943 0.8922 0.8931 0.4053 37.9273
1.1494 2.0 87 1.7744 0.4543 0.2046 0.3787 0.3803 0.8964 0.8918 0.8939 0.3972 36.2
1.7877 2.9885 130 2.8310 0.3344 0.1165 0.2735 0.2749 0.8579 0.8794 0.8676 0.3013 46.5273
2.83 4.0 174 2.7267 0.3339 0.1241 0.2813 0.2823 0.8636 0.8795 0.8709 0.3166 47.9182
2.6683 4.9885 217 2.6435 0.3296 0.1231 0.2791 0.2795 0.8478 0.8781 0.8615 0.3045 50.5455
2.5709 6.0 261 2.6193 0.3149 0.1164 0.2677 0.2679 0.828 0.8742 0.8488 0.2887 53.7818
2.6182 6.9885 304 2.6497 0.3028 0.1144 0.259 0.2598 0.7963 0.8706 0.8297 0.2619 59.4727
2.5553 8.0 348 2.6565 0.2994 0.1138 0.2581 0.2585 0.7895 0.8685 0.8248 0.2509 59.8909
2.6243 8.9885 391 2.6565 0.3001 0.1148 0.2576 0.2581 0.7911 0.8691 0.826 0.2547 60.1182
2.6506 10.0 435 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545
2.6527 10.9885 478 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545
2.6511 12.0 522 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545
2.612 12.9885 565 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545
2.6429 14.0 609 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545
2.6948 14.9885 652 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545
2.5908 16.0 696 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545
2.5995 16.9885 739 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545
2.5835 18.0 783 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545
2.6417 18.9885 826 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545
2.5769 19.7701 860 2.6565 0.3034 0.116 0.2608 0.2609 0.789 0.8697 0.8251 0.2559 60.4545

Framework versions

  • Transformers 4.44.0
  • Pytorch 2.4.0
  • Datasets 2.21.0
  • Tokenizers 0.19.1