Edit model card

N_roberta_agnews_padding0model

This model is a fine-tuned version of roberta-base on the ag_news dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5421
  • Accuracy: 0.9501

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: 20

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.1929 1.0 7500 0.2180 0.9363
0.1646 2.0 15000 0.2092 0.9455
0.1502 3.0 22500 0.2136 0.9478
0.1217 4.0 30000 0.2395 0.9476
0.1008 5.0 37500 0.2357 0.9501
0.0789 6.0 45000 0.3286 0.9420
0.0625 7.0 52500 0.3378 0.9439
0.0546 8.0 60000 0.4044 0.9443
0.0434 9.0 67500 0.4361 0.9412
0.0321 10.0 75000 0.4044 0.9453
0.0254 11.0 82500 0.4670 0.9455
0.0302 12.0 90000 0.4657 0.9438
0.0224 13.0 97500 0.4942 0.9432
0.0085 14.0 105000 0.5315 0.9449
0.0053 15.0 112500 0.5283 0.9455
0.01 16.0 120000 0.5004 0.9466
0.0061 17.0 127500 0.5430 0.9458
0.0042 18.0 135000 0.5116 0.9486
0.0034 19.0 142500 0.5379 0.9491
0.0022 20.0 150000 0.5421 0.9501

Framework versions

  • Transformers 4.33.2
  • Pytorch 2.0.1+cu117
  • Datasets 2.14.5
  • Tokenizers 0.13.3
Downloads last month
8
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 Realgon/N_roberta_agnews_padding0model

Finetuned
(1267)
this model

Dataset used to train Realgon/N_roberta_agnews_padding0model

Evaluation results