meat_calssify_fresh_crop_fixed_overlap_epoch100_V_0_12

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: 0.1999
• Accuracy: 0.9439

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
1.093	1.0	21	1.0798	0.4361
1.0819	2.0	42	1.0504	0.4704
1.0195	3.0	63	1.0107	0.4860
0.9205	4.0	84	0.9285	0.5514
0.879	5.0	105	0.8812	0.6044
0.757	6.0	126	0.8114	0.6324
0.6884	7.0	147	0.7328	0.6885
0.6399	8.0	168	0.7187	0.6978
0.5531	9.0	189	0.6771	0.7196
0.5187	10.0	210	0.6594	0.7134
0.5125	11.0	231	0.7660	0.6729
0.495	12.0	252	0.7215	0.7165
0.5014	13.0	273	0.5828	0.7570
0.3638	14.0	294	0.7056	0.7134
0.4493	15.0	315	0.7061	0.7383
0.4304	16.0	336	0.5031	0.7944
0.3223	17.0	357	0.5052	0.7975
0.3496	18.0	378	0.5136	0.8069
0.2498	19.0	399	0.5414	0.7944
0.3783	20.0	420	0.4276	0.8380
0.2768	21.0	441	0.4990	0.8100
0.2588	22.0	462	0.5184	0.8100
0.33	23.0	483	0.4037	0.8380
0.2418	24.0	504	0.4764	0.8100
0.2	25.0	525	0.3888	0.8505
0.1859	26.0	546	0.3868	0.8660
0.1804	27.0	567	0.5299	0.7944
0.1891	28.0	588	0.4448	0.8411
0.1837	29.0	609	0.4972	0.8349
0.209	30.0	630	0.4709	0.8380
0.1669	31.0	651	0.4084	0.8536
0.1474	32.0	672	0.4000	0.8785
0.1666	33.0	693	0.4109	0.8598
0.1657	34.0	714	0.3265	0.8910
0.1454	35.0	735	0.5221	0.8162
0.2093	36.0	756	0.6376	0.7944
0.1929	37.0	777	0.4007	0.8723
0.1393	38.0	798	0.3291	0.8879
0.1328	39.0	819	0.3766	0.8598
0.127	40.0	840	0.2965	0.9003
0.1325	41.0	861	0.3481	0.8723
0.118	42.0	882	0.3093	0.9065
0.1001	43.0	903	0.4232	0.8692
0.124	44.0	924	0.3761	0.8723
0.1159	45.0	945	0.3523	0.8910
0.129	46.0	966	0.3309	0.8785
0.1129	47.0	987	0.2915	0.9003
0.1043	48.0	1008	0.3259	0.8972
0.0986	49.0	1029	0.2627	0.9097
0.083	50.0	1050	0.3035	0.9034
0.0874	51.0	1071	0.3994	0.8629
0.0959	52.0	1092	0.2904	0.9065
0.0883	53.0	1113	0.2771	0.9128
0.0766	54.0	1134	0.2984	0.9128
0.0865	55.0	1155	0.3534	0.8941
0.0907	56.0	1176	0.3874	0.8723
0.0596	57.0	1197	0.2080	0.9283
0.0658	58.0	1218	0.3571	0.8879
0.0806	59.0	1239	0.3444	0.9003
0.0709	60.0	1260	0.3292	0.8972
0.0864	61.0	1281	0.3551	0.8816
0.0773	62.0	1302	0.2930	0.9159
0.0758	63.0	1323	0.2828	0.9221
0.0767	64.0	1344	0.2919	0.9065
0.0686	65.0	1365	0.2971	0.9065
0.0818	66.0	1386	0.3057	0.8972
0.0659	67.0	1407	0.2323	0.9221
0.0627	68.0	1428	0.3991	0.8754
0.0536	69.0	1449	0.2314	0.9221
0.2167	70.0	1470	0.2586	0.9346
0.0706	71.0	1491	0.2813	0.9315
0.0631	72.0	1512	0.2981	0.9034
0.0586	73.0	1533	0.2586	0.9283
0.0597	74.0	1554	0.3115	0.9097
0.0412	75.0	1575	0.2327	0.9315
0.0504	76.0	1596	0.2493	0.9408
0.0515	77.0	1617	0.2861	0.9283
0.0394	78.0	1638	0.2715	0.9128
0.0526	79.0	1659	0.2521	0.9190
0.043	80.0	1680	0.2421	0.9283
0.0466	81.0	1701	0.2918	0.9034
0.0418	82.0	1722	0.2956	0.9065
0.048	83.0	1743	0.2199	0.9283
0.0311	84.0	1764	0.2732	0.9128
0.0681	85.0	1785	0.2148	0.9346
0.0392	86.0	1806	0.2609	0.9252
0.0447	87.0	1827	0.2791	0.9346
0.0244	88.0	1848	0.2863	0.9221
0.0382	89.0	1869	0.2894	0.9190
0.0524	90.0	1890	0.1708	0.9408
0.0356	91.0	1911	0.2084	0.9221
0.0387	92.0	1932	0.2262	0.9377
0.0345	93.0	1953	0.2441	0.9377
0.0298	94.0	1974	0.2042	0.9408
0.0427	95.0	1995	0.1611	0.9533
0.043	96.0	2016	0.2175	0.9533
0.0241	97.0	2037	0.2445	0.9283
0.0416	98.0	2058	0.2236	0.9283
0.0311	99.0	2079	0.1943	0.9502
0.0352	100.0	2100	0.1999	0.9439

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.1
• Datasets 2.20.0
• Tokenizers 0.19.1