cvt-13-normal

This model is a fine-tuned version of microsoft/cvt-13 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 1.0613
Accuracy: 0.7790

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: 0.0005
train_batch_size: 142
eval_batch_size: 142
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 568
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	7	0.8652	0.7946
2.3799	2.0	14	0.8683	0.7912
2.4491	3.0	21	0.8807	0.7825
2.4491	4.0	28	0.9120	0.7851
2.3011	5.0	35	0.9865	0.7565
2.5444	6.0	42	0.9863	0.7643
2.5444	7.0	49	1.1580	0.7513
2.4127	8.0	56	1.1091	0.7383
2.8757	9.0	63	1.0644	0.7496
2.5231	10.0	70	1.0888	0.7400
2.5231	11.0	77	1.0668	0.7548
2.7538	12.0	84	1.0946	0.7435
2.7032	13.0	91	1.0676	0.7608
2.7032	14.0	98	1.0409	0.7426
2.4581	15.0	105	1.0679	0.7548
2.7023	16.0	112	1.0129	0.7487
2.7023	17.0	119	1.1501	0.7366
2.5456	18.0	126	1.0452	0.7426
2.7061	19.0	133	1.0034	0.7565
2.3491	20.0	140	1.0389	0.7574
2.3491	21.0	147	0.9999	0.7782
2.4926	22.0	154	1.0131	0.7652
2.5111	23.0	161	1.0940	0.7340
2.5111	24.0	168	1.0786	0.7582
2.3443	25.0	175	1.0768	0.7617
2.5738	26.0	182	0.9781	0.7782
2.5738	27.0	189	0.9955	0.7574
2.3528	28.0	196	1.0117	0.7669
2.599	29.0	203	1.0806	0.7660
2.3279	30.0	210	1.0101	0.7738
2.3279	31.0	217	1.0981	0.7617
2.5649	32.0	224	1.0185	0.7782
2.5432	33.0	231	1.1070	0.7591
2.5432	34.0	238	1.0705	0.7626
2.3521	35.0	245	1.0749	0.7574
2.5948	36.0	252	1.0508	0.7626
2.5948	37.0	259	1.0374	0.7712
2.3305	38.0	266	1.0249	0.7643
2.4833	39.0	273	1.0345	0.7712
2.1504	40.0	280	1.0252	0.7617
2.1504	41.0	287	1.0361	0.7574
2.4083	42.0	294	0.9939	0.7678
2.37	43.0	301	1.0186	0.7695
2.37	44.0	308	1.0861	0.7643
2.2043	45.0	315	1.0182	0.7643
2.3554	46.0	322	1.0584	0.7539
2.3554	47.0	329	1.0541	0.7617
2.1541	48.0	336	1.0967	0.7686
2.3739	49.0	343	1.1266	0.7721
2.1028	50.0	350	1.1116	0.7652
2.1028	51.0	357	1.0804	0.7643
2.3381	52.0	364	1.1142	0.7556
2.2902	53.0	371	1.1135	0.7652
2.2902	54.0	378	1.1024	0.7461
2.2452	55.0	385	1.0722	0.7626
2.4121	56.0	392	1.1089	0.7704
2.4121	57.0	399	1.0923	0.7548
2.2067	58.0	406	1.0811	0.7591
2.3894	59.0	413	1.1097	0.7634
2.2188	60.0	420	1.0988	0.7643
2.2188	61.0	427	1.0558	0.7686
2.2859	62.0	434	1.0569	0.7695
2.2293	63.0	441	1.1053	0.7643
2.2293	64.0	448	1.0962	0.7652
2.136	65.0	455	1.0505	0.7756
2.2507	66.0	462	1.0425	0.7799
2.2507	67.0	469	1.0703	0.7756
2.0269	68.0	476	1.0826	0.7695
2.2972	69.0	483	1.0569	0.7747
2.0192	70.0	490	1.0773	0.7695
2.0192	71.0	497	1.1000	0.7669
2.3668	72.0	504	1.1048	0.7712
2.1285	73.0	511	1.0883	0.7712
2.1285	74.0	518	1.0893	0.7738
2.0487	75.0	525	1.0644	0.7799
2.2508	76.0	532	1.0686	0.7764
2.2508	77.0	539	1.0759	0.7764
2.0141	78.0	546	1.0673	0.7756
2.1662	79.0	553	1.0610	0.7842
2.0567	80.0	560	1.0571	0.7851
2.0567	81.0	567	1.0682	0.7799
2.2602	82.0	574	1.0700	0.7782
2.3018	83.0	581	1.0703	0.7790
2.3018	84.0	588	1.0597	0.7825
2.0309	85.0	595	1.0560	0.7825
2.108	85.8	600	1.0613	0.7790

Framework versions

Transformers 4.47.1
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0

desarrolloasesoreslocales/cvt-13-normal

cvt-13-normal

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Seed this model