distilroberta-topic-classification_4

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

• Loss: 3.5428
• Acc: 0.7557

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 12345
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 16
• num_epochs: 20
• mixed_precision_training: Native AMP
• label_smoothing_factor: 0.5

Training results

Training Loss	Epoch	Step	Validation Loss	Acc
3.6261	1.0	564	3.6045	0.6610
3.5223	2.0	1128	3.5326	0.7111
3.4204	3.0	1692	3.5040	0.7322
3.3446	4.0	2256	3.4980	0.7400
3.2737	5.0	2820	3.4872	0.7539
3.2533	6.0	3384	3.4967	0.7555
3.2059	7.0	3948	3.5038	0.7613
3.1697	8.0	4512	3.5258	0.7537
3.1439	9.0	5076	3.5311	0.7573
3.1426	10.0	5640	3.5334	0.7544
3.1325	11.0	6204	3.5311	0.7562
3.1165	12.0	6768	3.5428	0.7557

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0