README.md

metadata

license: mit
base_model: roberta-base
tags:
  - generated_from_trainer
datasets:
  - ag_news
metrics:
  - accuracy
model-index:
  - name: N_roberta_agnews_padding70model
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: ag_news
          type: ag_news
          config: default
          split: test
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9465789473684211

N_roberta_agnews_padding70model

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: 0.5754
• Accuracy: 0.9466

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
0.201	1.0	7500	0.2029	0.9421
0.168	2.0	15000	0.2082	0.945
0.1533	3.0	22500	0.2343	0.9432
0.1208	4.0	30000	0.2381	0.9466
0.1071	5.0	37500	0.2468	0.9464
0.0831	6.0	45000	0.2775	0.9438
0.0758	7.0	52500	0.3080	0.9462
0.056	8.0	60000	0.3970	0.9436
0.0531	9.0	67500	0.3881	0.9401
0.037	10.0	75000	0.3956	0.9443
0.0309	11.0	82500	0.4551	0.9416
0.0257	12.0	90000	0.4521	0.9428
0.0287	13.0	97500	0.4650	0.9413
0.0121	14.0	105000	0.4888	0.9464
0.0116	15.0	112500	0.5071	0.9457
0.0085	16.0	120000	0.5249	0.9449
0.0107	17.0	127500	0.5244	0.9463
0.0031	18.0	135000	0.5597	0.9459
0.0041	19.0	142500	0.5615	0.9476
0.0029	20.0	150000	0.5754	0.9466

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3