meat_calssify_fresh_crop_fixed_overlap_epoch100_V_0_3

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.1880
• Accuracy: 0.9408

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: 64
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0748	1.0	21	1.0941	0.4019
1.056	2.0	42	1.0704	0.4517
1.0196	3.0	63	0.9922	0.5171
0.9144	4.0	84	0.9600	0.5140
0.9096	5.0	105	0.9206	0.5514
0.786	6.0	126	0.8006	0.6511
0.7149	7.0	147	0.7398	0.7196
0.6742	8.0	168	0.8100	0.6542
0.681	9.0	189	0.7297	0.6760
0.5929	10.0	210	0.7184	0.6854
0.5621	11.0	231	0.7011	0.7165
0.4628	12.0	252	0.6673	0.7196
0.4278	13.0	273	0.7029	0.7445
0.4525	14.0	294	0.6493	0.7477
0.3483	15.0	315	0.6969	0.7134
0.4328	16.0	336	0.5270	0.8006
0.3657	17.0	357	0.5653	0.7570
0.3047	18.0	378	0.4854	0.8131
0.2507	19.0	399	0.4555	0.8505
0.2468	20.0	420	0.5035	0.8131
0.2336	21.0	441	0.7171	0.7601
0.2954	22.0	462	0.4171	0.8536
0.2398	23.0	483	0.5465	0.7850
0.2538	24.0	504	0.5179	0.8069
0.21	25.0	525	0.3688	0.8723
0.1938	26.0	546	0.3997	0.8442
0.171	27.0	567	0.5068	0.8224
0.1983	28.0	588	0.4238	0.8380
0.1839	29.0	609	0.4431	0.8380
0.1977	30.0	630	0.3608	0.8598
0.1545	31.0	651	0.4898	0.8536
0.2214	32.0	672	0.5862	0.7850
0.185	33.0	693	0.3682	0.8785
0.1238	34.0	714	0.4300	0.8380
0.1424	35.0	735	0.5039	0.8287
0.1538	36.0	756	0.5649	0.8193
0.1806	37.0	777	0.3727	0.8505
0.1038	38.0	798	0.4984	0.8162
0.1241	39.0	819	0.3025	0.8941
0.1197	40.0	840	0.3038	0.8847
0.1288	41.0	861	0.5481	0.8100
0.1232	42.0	882	0.4011	0.8660
0.1308	43.0	903	0.3017	0.8910
0.0803	44.0	924	0.4368	0.8567
0.0893	45.0	945	0.3961	0.8660
0.1279	46.0	966	0.4143	0.8629
0.1105	47.0	987	0.3773	0.8598
0.0877	48.0	1008	0.3716	0.8816
0.0951	49.0	1029	0.3312	0.8847
0.0941	50.0	1050	0.2714	0.8910
0.073	51.0	1071	0.3475	0.8910
0.0878	52.0	1092	0.3918	0.8847
0.0898	53.0	1113	0.4729	0.8442
0.0849	54.0	1134	0.4245	0.8692
0.1619	55.0	1155	0.3289	0.9065
0.0838	56.0	1176	0.2787	0.9159
0.0767	57.0	1197	0.2738	0.9128
0.0815	58.0	1218	0.2729	0.9128
0.0747	59.0	1239	0.2036	0.9377
0.0629	60.0	1260	0.2615	0.9221
0.0561	61.0	1281	0.3424	0.8910
0.0666	62.0	1302	0.3222	0.8941
0.0759	63.0	1323	0.3462	0.9065
0.0548	64.0	1344	0.3463	0.8972
0.0607	65.0	1365	0.2171	0.9283
0.0796	66.0	1386	0.3879	0.8847
0.0651	67.0	1407	0.2649	0.9159
0.0615	68.0	1428	0.2469	0.9221
0.0495	69.0	1449	0.2899	0.9252
0.0511	70.0	1470	0.2891	0.9065
0.0487	71.0	1491	0.2990	0.9159
0.0593	72.0	1512	0.3046	0.9128
0.0484	73.0	1533	0.2865	0.9065
0.0534	74.0	1554	0.2614	0.9128
0.0446	75.0	1575	0.3311	0.8972
0.0478	76.0	1596	0.2580	0.9159
0.0335	77.0	1617	0.3392	0.9159
0.0436	78.0	1638	0.3400	0.9034
0.07	79.0	1659	0.3434	0.9034
0.0536	80.0	1680	0.3456	0.8972
0.0431	81.0	1701	0.2386	0.9408
0.0381	82.0	1722	0.2401	0.9346
0.0423	83.0	1743	0.2421	0.9346
0.0393	84.0	1764	0.1979	0.9439
0.0393	85.0	1785	0.2756	0.9190
0.0395	86.0	1806	0.3339	0.8972
0.031	87.0	1827	0.2471	0.9252
0.0227	88.0	1848	0.2182	0.9346
0.0392	89.0	1869	0.2732	0.9221
0.0536	90.0	1890	0.2579	0.9283
0.0426	91.0	1911	0.2062	0.9315
0.0344	92.0	1932	0.2209	0.9252
0.0333	93.0	1953	0.1584	0.9564
0.0338	94.0	1974	0.2976	0.9128
0.0391	95.0	1995	0.2420	0.9377
0.0302	96.0	2016	0.2694	0.9159
0.0268	97.0	2037	0.2610	0.9221
0.0402	98.0	2058	0.2952	0.9159
0.0172	99.0	2079	0.1870	0.9470
0.0241	100.0	2100	0.1880	0.9408

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0
• Datasets 2.19.2
• Tokenizers 0.19.1