distilroberta-rbm231k-ep20-op40-news_api_2p8k

This model is a fine-tuned version of judy93536/distilroberta-rbm231k-ep20-op40 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3991
• Accuracy: 0.8946

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: 2.153335054745316e-06
• 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
• lr_scheduler_warmup_ratio: 0.4
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	143	1.0982	0.3480
No log	2.0	286	1.0901	0.3480
No log	3.0	429	1.0754	0.3937
1.089	4.0	572	1.0446	0.4218
1.089	5.0	715	0.9983	0.4587
1.089	6.0	858	0.9316	0.5395
0.9811	7.0	1001	0.8240	0.6696
0.9811	8.0	1144	0.6669	0.7909
0.9811	9.0	1287	0.5055	0.8366
0.9811	10.0	1430	0.4123	0.8541
0.6112	11.0	1573	0.3531	0.8717
0.6112	12.0	1716	0.3399	0.8682
0.6112	13.0	1859	0.3304	0.8787
0.3218	14.0	2002	0.3270	0.8840
0.3218	15.0	2145	0.3201	0.8805
0.3218	16.0	2288	0.3276	0.8910
0.3218	17.0	2431	0.3326	0.8787
0.222	18.0	2574	0.3442	0.8928
0.222	19.0	2717	0.3456	0.8893
0.222	20.0	2860	0.3515	0.8893
0.1698	21.0	3003	0.3660	0.8893
0.1698	22.0	3146	0.3723	0.8928
0.1698	23.0	3289	0.3732	0.8928
0.1698	24.0	3432	0.3786	0.8928
0.1347	25.0	3575	0.3910	0.8963
0.1347	26.0	3718	0.3894	0.8963
0.1347	27.0	3861	0.3881	0.8981
0.1177	28.0	4004	0.3898	0.8981
0.1177	29.0	4147	0.3988	0.8946
0.1177	30.0	4290	0.3991	0.8946

Framework versions

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