beit-base-patch16-224-hasta-65-fold1

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

• Loss: 0.8281
• Accuracy: 0.6944

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• 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
No log	0.5714	1	1.1951	0.3333
No log	1.7143	3	1.2409	0.3056
No log	2.8571	5	1.2181	0.3056
No log	4.0	7	1.1308	0.5
No log	4.5714	8	1.1858	0.5278
1.092	5.7143	10	1.2798	0.3056
1.092	6.8571	12	1.0892	0.3611
1.092	8.0	14	1.2658	0.3889
1.092	8.5714	15	1.1603	0.5
1.092	9.7143	17	1.0786	0.4167
1.092	10.8571	19	1.2400	0.4167
0.9727	12.0	21	1.2929	0.3889
0.9727	12.5714	22	1.1613	0.3889
0.9727	13.7143	24	0.9838	0.6111
0.9727	14.8571	26	1.2316	0.3889
0.9727	16.0	28	1.1520	0.4722
0.9727	16.5714	29	1.0345	0.5556
0.8844	17.7143	31	1.0000	0.5
0.8844	18.8571	33	0.9933	0.5278
0.8844	20.0	35	1.0512	0.5556
0.8844	20.5714	36	0.9950	0.5556
0.8844	21.7143	38	0.9621	0.4722
0.7447	22.8571	40	0.8812	0.5278
0.7447	24.0	42	1.0244	0.5833
0.7447	24.5714	43	1.0124	0.5833
0.7447	25.7143	45	0.8908	0.6389
0.7447	26.8571	47	0.8185	0.5833
0.7447	28.0	49	0.9409	0.5556
0.6176	28.5714	50	1.0401	0.5556
0.6176	29.7143	52	1.0989	0.5556
0.6176	30.8571	54	0.9102	0.5833
0.6176	32.0	56	0.8855	0.5833
0.6176	32.5714	57	0.8974	0.5556
0.6176	33.7143	59	0.9419	0.6111
0.4929	34.8571	61	0.9471	0.5833
0.4929	36.0	63	0.8609	0.5833
0.4929	36.5714	64	0.8558	0.5833
0.4929	37.7143	66	0.8449	0.5556
0.4929	38.8571	68	0.8136	0.6667
0.463	40.0	70	0.8281	0.6944
0.463	40.5714	71	0.8227	0.6944
0.463	41.7143	73	0.8323	0.5833
0.463	42.8571	75	0.8436	0.5833
0.463	44.0	77	0.8390	0.5833
0.463	44.5714	78	0.8580	0.6111
0.3995	45.7143	80	0.9375	0.6111
0.3995	46.8571	82	0.9897	0.5556
0.3995	48.0	84	0.9785	0.5556
0.3995	48.5714	85	0.9336	0.6389
0.3995	49.7143	87	0.8504	0.6389
0.3995	50.8571	89	0.8450	0.6667
0.3697	52.0	91	0.8531	0.6389
0.3697	52.5714	92	0.8728	0.6389
0.3697	53.7143	94	0.9076	0.6667
0.3697	54.8571	96	0.9175	0.6389
0.3697	56.0	98	0.9145	0.5833
0.3697	56.5714	99	0.9119	0.5833
0.3259	57.1429	100	0.9102	0.5833

Framework versions

• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1