smids_5x_deit_small_rms_001_fold4

This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.5824
• Accuracy: 0.78

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: 0.001
• train_batch_size: 32
• eval_batch_size: 32
• 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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9095	1.0	375	0.8145	0.53
0.8394	2.0	750	0.8071	0.565
0.7959	3.0	1125	0.7903	0.6133
0.7869	4.0	1500	0.7586	0.6367
0.7989	5.0	1875	0.7552	0.6
0.8124	6.0	2250	0.7289	0.66
0.7708	7.0	2625	0.7042	0.6733
0.7586	8.0	3000	0.7504	0.6583
0.6986	9.0	3375	0.6527	0.6983
0.6979	10.0	3750	0.6544	0.7017
0.682	11.0	4125	0.6621	0.7117
0.6815	12.0	4500	0.6293	0.7067
0.6311	13.0	4875	0.6466	0.7033
0.743	14.0	5250	0.5967	0.7383
0.6884	15.0	5625	0.5874	0.7533
0.6214	16.0	6000	0.5678	0.7567
0.6379	17.0	6375	0.6145	0.7267
0.5615	18.0	6750	0.5793	0.7417
0.5825	19.0	7125	0.5647	0.76
0.5806	20.0	7500	0.5298	0.7617
0.5732	21.0	7875	0.6497	0.7117
0.4981	22.0	8250	0.6229	0.7283
0.5878	23.0	8625	0.5476	0.77
0.5732	24.0	9000	0.5431	0.7783
0.5633	25.0	9375	0.5734	0.7617
0.5704	26.0	9750	0.5553	0.7683
0.537	27.0	10125	0.5504	0.7733
0.4571	28.0	10500	0.5331	0.7783
0.5264	29.0	10875	0.5680	0.7633
0.6141	30.0	11250	0.5510	0.765
0.5469	31.0	11625	0.5500	0.7933
0.4915	32.0	12000	0.5001	0.785
0.5227	33.0	12375	0.5958	0.7783
0.4961	34.0	12750	0.5665	0.78
0.4306	35.0	13125	0.5345	0.7683
0.461	36.0	13500	0.5456	0.7683
0.5254	37.0	13875	0.5228	0.78
0.4633	38.0	14250	0.5026	0.7917
0.4546	39.0	14625	0.5577	0.7633
0.4842	40.0	15000	0.5245	0.78
0.4453	41.0	15375	0.5350	0.785
0.3943	42.0	15750	0.5494	0.7867
0.4031	43.0	16125	0.5697	0.7833
0.3729	44.0	16500	0.5326	0.7933
0.3744	45.0	16875	0.5371	0.7817
0.4535	46.0	17250	0.5557	0.7817
0.4267	47.0	17625	0.5568	0.7767
0.372	48.0	18000	0.5642	0.77
0.3734	49.0	18375	0.5737	0.785
0.4125	50.0	18750	0.5824	0.78

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2