db_mc2_29.1.3nm

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

• Loss: 0.2933
• Accuracy: 0.9612
• Balanced Accuracy: 0.9651
• F1 Weighted: 0.9611
• Precision Weighted: 0.9612
• Recall Weighted: 0.9612
• F1 Macro: 0.9645
• Precision Macro: 0.9641
• Recall Macro: 0.9651
• F1 Min: 0.8367
• N Below 80: 0
• N Errors: 631
• Conf Err Rate: 0.5578

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: 3.023887626013629e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 0.09
• num_epochs: 15
• label_smoothing_factor: 0.01

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Balanced Accuracy	F1 Weighted	Precision Weighted	Recall Weighted	F1 Macro	Precision Macro	Recall Macro	F1 Min	N Below 80	N Errors	Conf Err Rate
1.5451	1.0	1439	1.2228	0.7441	0.7423	0.7349	0.7554	0.7441	0.7349	0.7606	0.7423	0.2641	54	4158	0.0012
0.4941	2.0	2878	0.4246	0.9083	0.9143	0.9076	0.9085	0.9083	0.9138	0.9149	0.9143	0.7136	9	1489	0.1296
0.2990	3.0	4317	0.3443	0.9354	0.9406	0.9353	0.9362	0.9354	0.9402	0.9408	0.9406	0.7573	3	1049	0.2316
0.2296	4.0	5756	0.3214	0.9431	0.9489	0.9431	0.9441	0.9431	0.9477	0.9475	0.9489	0.7844	1	924	0.3019
0.1778	5.0	7195	0.3132	0.9471	0.9516	0.9471	0.9481	0.9471	0.9510	0.9515	0.9516	0.8196	0	859	0.3551
0.1446	6.0	8634	0.3038	0.9523	0.9572	0.9522	0.9526	0.9523	0.9563	0.9558	0.9572	0.8018	0	775	0.4090
0.1341	7.0	10073	0.3081	0.9541	0.9596	0.9540	0.9546	0.9541	0.9583	0.9575	0.9596	0.8061	0	746	0.4799
0.1183	8.0	11512	0.2990	0.9554	0.9599	0.9553	0.9557	0.9554	0.9594	0.9593	0.9599	0.8159	0	725	0.4717
0.1101	9.0	12951	0.2999	0.9566	0.9609	0.9565	0.9568	0.9566	0.9603	0.9601	0.9609	0.8314	0	706	0.5
0.1068	10.0	14390	0.2954	0.9586	0.9623	0.9585	0.9587	0.9586	0.9621	0.9623	0.9623	0.8368	0	673	0.5126
0.1073	11.0	15829	0.2955	0.9593	0.9633	0.9592	0.9594	0.9593	0.9628	0.9627	0.9633	0.8331	0	662	0.5332
0.1027	12.0	17268	0.2964	0.9594	0.9634	0.9594	0.9596	0.9594	0.9628	0.9626	0.9634	0.8363	0	659	0.5493
0.1030	13.0	18707	0.2937	0.9605	0.9643	0.9604	0.9606	0.9605	0.9637	0.9634	0.9643	0.8406	0	642	0.5467
0.1022	14.0	20146	0.2933	0.9612	0.9651	0.9611	0.9612	0.9612	0.9645	0.9641	0.9651	0.8367	0	631	0.5578
0.1021	15.0	21585	0.2929	0.9612	0.9650	0.9611	0.9612	0.9612	0.9645	0.9641	0.9650	0.8372	0	631	0.5547

Framework versions

• Transformers 5.15.1
• Pytorch 2.11.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2