smids_10x_beit_large_sgd_00001_fold3

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

• Loss: 0.8246
• Accuracy: 0.6317

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: 1e-05
• 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
1.2284	1.0	750	1.2241	0.3517
1.1457	2.0	1500	1.1930	0.365
1.1396	3.0	2250	1.1661	0.3783
1.0897	4.0	3000	1.1425	0.385
1.025	5.0	3750	1.1215	0.3883
1.0158	6.0	4500	1.1022	0.3883
0.9975	7.0	5250	1.0842	0.4017
1.0278	8.0	6000	1.0673	0.4067
0.9784	9.0	6750	1.0514	0.4133
0.9157	10.0	7500	1.0366	0.4317
0.9554	11.0	8250	1.0228	0.4467
0.8899	12.0	9000	1.0096	0.4667
0.9379	13.0	9750	0.9973	0.4767
0.944	14.0	10500	0.9856	0.4867
0.9071	15.0	11250	0.9745	0.4983
0.8922	16.0	12000	0.9641	0.505
0.8643	17.0	12750	0.9544	0.5133
0.8278	18.0	13500	0.9449	0.52
0.9039	19.0	14250	0.9361	0.5317
0.8559	20.0	15000	0.9279	0.5383
0.8179	21.0	15750	0.9199	0.545
0.8248	22.0	16500	0.9124	0.56
0.8379	23.0	17250	0.9052	0.56
0.864	24.0	18000	0.8985	0.565
0.8458	25.0	18750	0.8922	0.575
0.8014	26.0	19500	0.8861	0.5783
0.7589	27.0	20250	0.8805	0.5883
0.8089	28.0	21000	0.8752	0.595
0.8337	29.0	21750	0.8701	0.5983
0.7734	30.0	22500	0.8654	0.6033
0.7463	31.0	23250	0.8610	0.6033
0.7746	32.0	24000	0.8569	0.6067
0.8126	33.0	24750	0.8532	0.6117
0.7894	34.0	25500	0.8496	0.615
0.7634	35.0	26250	0.8463	0.615
0.7765	36.0	27000	0.8433	0.6167
0.8136	37.0	27750	0.8405	0.6217
0.8117	38.0	28500	0.8380	0.6217
0.7707	39.0	29250	0.8357	0.6217
0.7678	40.0	30000	0.8337	0.6267
0.7823	41.0	30750	0.8319	0.6283
0.7728	42.0	31500	0.8303	0.63
0.7705	43.0	32250	0.8289	0.6283
0.7342	44.0	33000	0.8277	0.6283
0.7107	45.0	33750	0.8267	0.6283
0.7263	46.0	34500	0.8259	0.63
0.7101	47.0	35250	0.8253	0.63
0.7724	48.0	36000	0.8249	0.6317
0.7714	49.0	36750	0.8247	0.6317
0.7461	50.0	37500	0.8246	0.6317

Framework versions

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