README.md

metadata

library_name: transformers
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: vit-base-patch16-224-in21k-finetuned-papsmear
    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: 0.8897058823529411

vit-base-patch16-224-in21k-finetuned-papsmear

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.3853
• Accuracy: 0.8897

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.9231	9	1.7589	0.2426
1.7862	1.9487	19	1.5880	0.3824
1.6727	2.9744	29	1.4212	0.4265
1.5102	4.0	39	1.2241	0.5809
1.3247	4.9231	48	1.0906	0.6103
1.1047	5.9487	58	0.9747	0.6765
0.9405	6.9744	68	0.8745	0.7426
0.823	8.0	78	0.7833	0.7426
0.7244	8.9231	87	0.7160	0.7794
0.6367	9.9487	97	0.7328	0.7794
0.5537	10.9744	107	0.6573	0.7868
0.484	12.0	117	0.5988	0.8088
0.4642	12.9231	126	0.6268	0.7941
0.4166	13.9487	136	0.6549	0.7794
0.4106	14.9744	146	0.5330	0.8529
0.3947	16.0	156	0.5134	0.8382
0.3469	16.9231	165	0.5879	0.7794
0.3151	17.9487	175	0.5683	0.8382
0.2946	18.9744	185	0.5383	0.8162
0.2927	20.0	195	0.5682	0.8162
0.2879	20.9231	204	0.4722	0.8603
0.2512	21.9487	214	0.4806	0.8456
0.2633	22.9744	224	0.4713	0.8456
0.2286	24.0	234	0.5167	0.8382
0.2265	24.9231	243	0.3886	0.8824
0.2107	25.9487	253	0.4396	0.8676
0.2044	26.9744	263	0.4734	0.8456
0.1925	28.0	273	0.4606	0.8529
0.1866	28.9231	282	0.5061	0.8309
0.1928	29.9487	292	0.4202	0.8824
0.1907	30.9744	302	0.5120	0.8309
0.1631	32.0	312	0.4165	0.8676
0.1654	32.9231	321	0.4600	0.8676
0.154	33.9487	331	0.3834	0.8971
0.1459	34.9744	341	0.3686	0.8897
0.1452	36.0	351	0.4174	0.8676
0.1548	36.9231	360	0.3791	0.9044
0.1395	37.9487	370	0.4512	0.8529
0.1333	38.9744	380	0.3775	0.8897
0.1236	40.0	390	0.3666	0.8971
0.1236	40.9231	399	0.3892	0.8971
0.1314	41.9487	409	0.3832	0.8897
0.1322	42.9744	419	0.3919	0.8824
0.1156	44.0	429	0.3699	0.8971
0.1222	44.9231	438	0.3828	0.8971
0.1254	45.9487	448	0.3853	0.8897
0.1129	46.1538	450	0.3853	0.8897

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1