Hugging Face's logo

RMWeerasinghe
/
long-t5-tglobal-base-boardpapers-4096

Summarization
Transformers
Safetensors
longt5
text2text-generation
Generated from Trainer
Inference Endpoints
Model card Files Community
long-t5-tglobal-base-boardpapers-4096 / README.md
RMWeerasinghe's picture
RMWeerasinghe
Training complete
6e6d386 verified
preview code
|
raw
history blame
No virus
3.44 kB
metadata
license: apache-2.0
base_model: RMWeerasinghe/long-t5-tglobal-base-finetuned-govReport-4096
tags:
  - summarization
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: long-t5-tglobal-base-boardpapers-4096
    results: []

long-t5-tglobal-base-boardpapers-4096

This model is a fine-tuned version of RMWeerasinghe/long-t5-tglobal-base-finetuned-govReport-4096 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5617
  • Rouge1: 0.0743
  • Rouge2: 0.0398
  • Rougel: 0.0589
  • Rougelsum: 0.0703

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: 4e-05
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 32
  • 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
No log 0.67 1 0.6654 0.0514 0.0197 0.0386 0.0477
No log 2.0 3 0.6378 0.0667 0.0309 0.0512 0.0596
No log 2.67 4 0.6293 0.0646 0.0274 0.0515 0.0619
No log 4.0 6 0.6128 0.0706 0.0377 0.0566 0.067
No log 4.67 7 0.6049 0.0706 0.0377 0.0566 0.067
No log 6.0 9 0.5935 0.0706 0.0377 0.0566 0.067
No log 6.67 10 0.5891 0.0718 0.0385 0.0578 0.067
No log 8.0 12 0.5815 0.0743 0.0398 0.0589 0.0703
No log 8.67 13 0.5785 0.0743 0.0398 0.0589 0.0703
No log 10.0 15 0.5742 0.0743 0.0398 0.0589 0.0703
No log 10.67 16 0.5724 0.0743 0.0398 0.0589 0.0703
No log 12.0 18 0.5694 0.0743 0.0398 0.0589 0.0703
No log 12.67 19 0.5681 0.0743 0.0398 0.0589 0.0703
0.7929 14.0 21 0.5661 0.0743 0.0398 0.0589 0.0703
0.7929 14.67 22 0.5652 0.0743 0.0398 0.0589 0.0703
0.7929 16.0 24 0.5636 0.0743 0.0398 0.0589 0.0703
0.7929 16.67 25 0.5630 0.0743 0.0398 0.0589 0.0703
0.7929 18.0 27 0.5621 0.0743 0.0398 0.0589 0.0703
0.7929 18.67 28 0.5619 0.0743 0.0398 0.0589 0.0703
0.7929 20.0 30 0.5617 0.0743 0.0398 0.0589 0.0703

Framework versions

  • Transformers 4.37.0
  • Pytorch 2.1.2
  • Datasets 2.17.0
  • Tokenizers 0.15.1