metadata

license: apache-2.0
base_model: google/vit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - precision
  - recall
model-index:
  - name: vit-base-patch16-224
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: validation
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8566666666666667
          - name: Precision
            type: precision
            value: 0.8522571872571872
          - name: Recall
            type: recall
            value: 0.8566666666666667

vit-base-patch16-224

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

Loss: 0.4410
Accuracy: 0.8567
Precision: 0.8523
Recall: 0.8567
F1 Score: 0.8517

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: 64
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 256
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 Score
No log	1.0	4	0.5841	0.7333	0.6770	0.7333	0.6479
No log	2.0	8	0.5727	0.7333	0.5378	0.7333	0.6205
No log	3.0	12	0.6089	0.7208	0.7222	0.7208	0.7215
No log	4.0	16	0.5332	0.7458	0.7205	0.7458	0.6727
No log	5.0	20	0.5314	0.7625	0.7410	0.7625	0.7416
No log	6.0	24	0.5284	0.7583	0.7486	0.7583	0.6959
No log	7.0	28	0.5220	0.775	0.7700	0.775	0.7286
0.5564	8.0	32	0.5204	0.7833	0.7740	0.7833	0.7481
0.5564	9.0	36	0.5044	0.7708	0.7616	0.7708	0.7650
0.5564	10.0	40	0.4845	0.8125	0.8051	0.8125	0.7941
0.5564	11.0	44	0.4921	0.7833	0.7726	0.7833	0.7757
0.5564	12.0	48	0.4792	0.8167	0.8098	0.8167	0.7996
0.5564	13.0	52	0.4825	0.8	0.7889	0.8	0.7901
0.5564	14.0	56	0.4987	0.8083	0.7989	0.8083	0.8002
0.3176	15.0	60	0.4970	0.8208	0.8144	0.8208	0.8050
0.3176	16.0	64	0.5076	0.8083	0.7983	0.8083	0.7923
0.3176	17.0	68	0.5227	0.8083	0.7979	0.8083	0.7941
0.3176	18.0	72	0.5132	0.8042	0.7928	0.8042	0.7905
0.3176	19.0	76	0.5081	0.8167	0.8087	0.8167	0.8014
0.3176	20.0	80	0.5140	0.8292	0.8220	0.8292	0.8187
0.3176	21.0	84	0.5392	0.8125	0.8032	0.8125	0.7977
0.3176	22.0	88	0.5175	0.7958	0.7829	0.7958	0.7815
0.1778	23.0	92	0.5109	0.8125	0.8032	0.8125	0.7977
0.1778	24.0	96	0.4961	0.8292	0.8217	0.8292	0.8213
0.1778	25.0	100	0.5251	0.8083	0.7979	0.8083	0.7941
0.1778	26.0	104	0.5192	0.8167	0.8075	0.8167	0.8046
0.1778	27.0	108	0.5030	0.8333	0.8274	0.8333	0.8286
0.1778	28.0	112	0.5031	0.8375	0.8310	0.8375	0.8300
0.1778	29.0	116	0.5164	0.8208	0.8127	0.8208	0.8083
0.1109	30.0	120	0.5192	0.8208	0.8127	0.8208	0.8083

Framework versions

Transformers 4.33.3
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3