README.md

metadata

license: apache-2.0
base_model: microsoft/beit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: beit-base-patch16-224-hasta-65-fold4
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6944444444444444

beit-base-patch16-224-hasta-65-fold4

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: 0.7415
• 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	1.4459	0.3333
No log	1.7143	3	1.1743	0.3889
No log	2.8571	5	1.1216	0.3056
No log	4.0	7	1.1048	0.2778
No log	4.5714	8	1.0513	0.5
1.1273	5.7143	10	1.1055	0.3333
1.1273	6.8571	12	1.0529	0.4444
1.1273	8.0	14	1.0445	0.4722
1.1273	8.5714	15	1.0336	0.4722
1.1273	9.7143	17	0.9757	0.4444
1.1273	10.8571	19	0.9972	0.4444
0.9616	12.0	21	0.9694	0.5278
0.9616	12.5714	22	0.9377	0.4722
0.9616	13.7143	24	0.8975	0.5556
0.9616	14.8571	26	0.9970	0.4444
0.9616	16.0	28	0.9322	0.5833
0.9616	16.5714	29	0.9820	0.5278
0.8463	17.7143	31	1.1023	0.5
0.8463	18.8571	33	1.1089	0.5
0.8463	20.0	35	0.9417	0.5556
0.8463	20.5714	36	0.8424	0.5833
0.8463	21.7143	38	0.8668	0.6111
0.7082	22.8571	40	0.9767	0.5556
0.7082	24.0	42	0.8743	0.6389
0.7082	24.5714	43	0.7945	0.6389
0.7082	25.7143	45	0.9246	0.5278
0.7082	26.8571	47	1.2622	0.5833
0.7082	28.0	49	0.7754	0.5278
0.6413	28.5714	50	0.7375	0.5833
0.6413	29.7143	52	1.0095	0.5556
0.6413	30.8571	54	1.0806	0.5833
0.6413	32.0	56	0.7415	0.6944
0.6413	32.5714	57	0.7523	0.6944
0.6413	33.7143	59	0.9506	0.6111
0.5256	34.8571	61	0.9487	0.6667
0.5256	36.0	63	0.8945	0.6111
0.5256	36.5714	64	0.9073	0.6111
0.5256	37.7143	66	0.9394	0.6389
0.5256	38.8571	68	0.9062	0.6389
0.4509	40.0	70	0.8908	0.6111
0.4509	40.5714	71	0.8960	0.6111
0.4509	41.7143	73	0.9506	0.6389
0.4509	42.8571	75	1.0018	0.6111
0.4509	44.0	77	0.9852	0.6667
0.4509	44.5714	78	1.0045	0.6667
0.3865	45.7143	80	1.0984	0.5556
0.3865	46.8571	82	1.1893	0.5556
0.3865	48.0	84	1.2066	0.5278
0.3865	48.5714	85	1.1625	0.5556
0.3865	49.7143	87	1.0753	0.6111
0.3865	50.8571	89	1.0610	0.6111
0.3497	52.0	91	1.0844	0.5833
0.3497	52.5714	92	1.1055	0.5556
0.3497	53.7143	94	1.1122	0.5556
0.3497	54.8571	96	1.1042	0.5833
0.3497	56.0	98	1.0855	0.5556
0.3497	56.5714	99	1.0785	0.5833
0.3196	57.1429	100	1.0751	0.5833

Framework versions

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