Edit model card

T5_base_title_v4

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

  • Loss: 1.6697
  • Rouge1: 0.4305
  • Rouge2: 0.2304
  • Rougel: 0.3728
  • Rougelsum: 0.3729
  • Gen Len: 16.6586

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: 15
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
1.9653 1.0 2019 1.7927 0.4092 0.2145 0.3528 0.3528 16.6021
1.828 2.0 4038 1.7374 0.4148 0.217 0.3557 0.3558 16.7601
1.7597 3.0 6057 1.7053 0.4183 0.2199 0.3595 0.3594 16.8878
1.6787 4.0 8076 1.6875 0.4221 0.224 0.3649 0.3647 16.6098
1.6361 5.0 10095 1.6730 0.4227 0.2229 0.3655 0.3657 16.6044
1.6032 6.0 12114 1.6679 0.4266 0.227 0.3696 0.3697 16.4617
1.5701 7.0 14133 1.6657 0.4265 0.2273 0.3694 0.3692 16.4184
1.5359 8.0 16152 1.6677 0.4273 0.2274 0.3695 0.3695 16.5704
1.5136 9.0 18171 1.6639 0.4271 0.2278 0.3697 0.3697 16.5989
1.4776 10.0 20190 1.6641 0.4291 0.2297 0.3723 0.3722 16.5137
1.4507 11.0 22209 1.6650 0.4307 0.2303 0.372 0.3718 16.5868
1.437 12.0 24228 1.6654 0.4277 0.2274 0.3711 0.3711 16.7277
1.4428 13.0 26247 1.6689 0.4296 0.2287 0.3714 0.3715 16.7078
1.4183 14.0 28266 1.6697 0.4307 0.2301 0.3726 0.3725 16.6979
1.4244 15.0 30285 1.6697 0.4305 0.2304 0.3728 0.3729 16.6586

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.1.2
  • Datasets 2.1.0
  • Tokenizers 0.15.1
Downloads last month
6
Safetensors
Model size
223M 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 Swarnava/T5_base_title_v4

Base model

google-t5/t5-base
Finetuned
(380)
this model