fine-tuned-BioBART-12-epochs-1024-input-128-output

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

  • Loss: 1.5051
  • Rouge1: 0.1576
  • Rouge2: 0.0366
  • Rougel: 0.108
  • Rougelsum: 0.1085
  • Gen Len: 34.29

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: 0.0001
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 12

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
No log 1.0 151 4.4833 0.0206 0.0051 0.0196 0.02 7.18
No log 2.0 302 1.8298 0.1027 0.0249 0.0865 0.0866 26.29
No log 3.0 453 1.6450 0.0738 0.0169 0.0564 0.0568 20.31
3.8819 4.0 604 1.5676 0.1469 0.0336 0.1162 0.1163 33.72
3.8819 5.0 755 1.5271 0.1739 0.0342 0.1442 0.1441 29.99
3.8819 6.0 906 1.5005 0.1616 0.0337 0.1275 0.1286 36.78
1.1888 7.0 1057 1.4912 0.1618 0.0417 0.121 0.1212 41.89
1.1888 8.0 1208 1.4854 0.1293 0.0396 0.0905 0.0911 36.59
1.1888 9.0 1359 1.4949 0.1451 0.0337 0.1153 0.1158 27.18
0.8646 10.0 1510 1.4936 0.1743 0.0355 0.1292 0.1301 34.24
0.8646 11.0 1661 1.5060 0.1639 0.0367 0.1233 0.124 30.56
0.8646 12.0 1812 1.5051 0.1576 0.0366 0.108 0.1085 34.29

Framework versions

  • Transformers 4.36.2
  • Pytorch 1.12.1+cu113
  • Datasets 2.16.1
  • Tokenizers 0.15.0
Downloads last month
9
Safetensors
Model size
139M 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 tanatapanun/fine-tuned-BioBART-12-epochs-1024-input-128-output

Finetuned
(12)
this model