metadata

license: apache-2.0
base_model: google/vit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: vit-base-patch16-224-U8-40c
    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: 0.8235294117647058

vit-base-patch16-224-U8-40c

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: 0.5609
Accuracy: 0.8235

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: 0.05
num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.3495	1.0	20	1.3142	0.4706
1.1689	2.0	40	1.1153	0.5686
0.8673	3.0	60	0.8498	0.6667
0.5847	4.0	80	0.7220	0.7843
0.4029	5.0	100	0.8654	0.6275
0.2562	6.0	120	0.5609	0.8235
0.2352	7.0	140	0.7272	0.7843
0.2131	8.0	160	0.7581	0.7255
0.1616	9.0	180	0.5437	0.8235
0.1266	10.0	200	0.6345	0.8039
0.1557	11.0	220	0.8280	0.7647
0.0871	12.0	240	0.9016	0.7059
0.0879	13.0	260	0.8099	0.7647
0.0844	14.0	280	0.8791	0.7255
0.0865	15.0	300	0.9713	0.7843
0.1005	16.0	320	0.9966	0.7843
0.0718	17.0	340	1.0468	0.7647
0.0591	18.0	360	0.9471	0.7843
0.0641	19.0	380	0.9905	0.7451
0.0542	20.0	400	1.0300	0.7451
0.0813	21.0	420	1.0330	0.7647
0.059	22.0	440	0.9995	0.7647
0.0679	23.0	460	0.9327	0.7451
0.0611	24.0	480	1.0073	0.7647
0.0694	25.0	500	0.9348	0.7647
0.0454	26.0	520	0.8551	0.7843
0.0536	27.0	540	0.9782	0.7647
0.0429	28.0	560	0.9203	0.7843
0.0386	29.0	580	0.8732	0.8039
0.0433	30.0	600	0.9376	0.7647
0.0353	31.0	620	0.8532	0.7843
0.0332	32.0	640	0.9123	0.8039
0.0405	33.0	660	0.9603	0.8039
0.0423	34.0	680	0.9424	0.8039
0.0383	35.0	700	0.9687	0.8235
0.0245	36.0	720	0.9509	0.8235
0.0309	37.0	740	0.8950	0.8235
0.026	38.0	760	0.9082	0.8039
0.0192	39.0	780	0.8859	0.8235
0.0322	40.0	800	0.8968	0.8235

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu118
Datasets 2.16.1
Tokenizers 0.15.0