rare-breeze-233

This model is a fine-tuned version of facebook/convnextv2-tiny-1k-224 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0991
• Accuracy: 0.9695
• Precision: 0.9711
• Recall: 0.9695
• F1: 0.9697
• Roc Auc: 0.9974

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: 0.0001
• train_batch_size: 256
• eval_batch_size: 256
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Roc Auc
1.3478	1.0	15	1.2915	0.3594	0.5281	0.3594	0.3551	0.6948
1.2146	2.0	30	1.1017	0.5469	0.5974	0.5469	0.5562	0.7841
1.0294	3.0	45	0.9505	0.5758	0.5723	0.5758	0.5491	0.8132
0.9032	4.0	60	0.8093	0.5961	0.5979	0.5961	0.5964	0.8319
0.7706	5.0	75	0.6846	0.6430	0.6440	0.6430	0.6361	0.8620
0.6483	6.0	90	0.6121	0.5891	0.7346	0.5891	0.5746	0.8751
0.5611	7.0	105	0.5871	0.6016	0.7943	0.6016	0.5815	0.9205
0.4989	8.0	120	0.5180	0.6844	0.7926	0.6844	0.6850	0.9371
0.4009	9.0	135	0.3438	0.7703	0.7888	0.7703	0.7715	0.9464
0.4339	10.0	150	0.4494	0.8242	0.8259	0.8242	0.8239	0.9506
0.2688	11.0	165	0.3047	0.8297	0.8346	0.8297	0.8239	0.9689
0.1994	12.0	180	0.3227	0.7867	0.8292	0.7867	0.7882	0.9604
0.1293	13.0	195	0.2106	0.8664	0.8906	0.8664	0.8674	0.9860
0.0769	14.0	210	0.2103	0.9086	0.9113	0.9086	0.9078	0.9896
0.0708	15.0	225	0.1683	0.9352	0.9367	0.9352	0.9349	0.9932
0.049	16.0	240	0.1728	0.925	0.9290	0.925	0.9238	0.9931
0.0611	17.0	255	0.1687	0.9031	0.9122	0.9031	0.9047	0.9906
0.0344	18.0	270	0.1366	0.9539	0.9556	0.9539	0.9540	0.9949
0.0255	19.0	285	0.1696	0.9375	0.9408	0.9375	0.9378	0.9932
0.0207	20.0	300	0.1629	0.9422	0.9442	0.9422	0.9418	0.9957
0.0297	21.0	315	0.1349	0.9625	0.9629	0.9625	0.9625	0.9943
0.0125	22.0	330	0.1054	0.9727	0.9731	0.9727	0.9727	0.9967
0.0092	23.0	345	0.0892	0.9734	0.9734	0.9734	0.9734	0.9968
0.0116	24.0	360	0.0694	0.9703	0.9707	0.9703	0.9704	0.9978
0.0103	25.0	375	0.0991	0.9695	0.9711	0.9695	0.9697	0.9974

Framework versions

• Transformers 4.52.3
• Pytorch 2.7.0+cpu
• Datasets 3.6.0
• Tokenizers 0.21.0