metadata

license: apache-2.0
base_model: google/vit-base-patch16-224
tags:
  - image-classification
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: vit-lr-poly
    results: []

vit-lr-poly

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

Loss: 0.4501
Accuracy: 0.8488
Precision: 0.8418
Recall: 0.8488
F1: 0.8427

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: 0.0001
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: polynomial
num_epochs: 100
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.5905	0.31	100	0.6207	0.7788	0.7719	0.7788	0.7700
0.5605	0.62	200	0.7325	0.7621	0.7503	0.7621	0.7152
0.7068	0.93	300	0.5869	0.7920	0.8184	0.7920	0.7952
0.3773	1.25	400	0.5412	0.7854	0.8200	0.7854	0.7964
0.3501	1.56	500	0.5548	0.8214	0.8133	0.8214	0.8029
0.31	1.87	600	0.6007	0.7881	0.8345	0.7881	0.7906
0.1492	2.18	700	0.4845	0.8370	0.8433	0.8370	0.8340
0.185	2.49	800	0.4501	0.8488	0.8418	0.8488	0.8427
0.2438	2.8	900	0.4976	0.8440	0.8412	0.8440	0.8338
0.0604	3.12	1000	0.5850	0.8408	0.8425	0.8408	0.8405
0.0545	3.43	1100	0.5685	0.8492	0.8476	0.8492	0.8445
0.0719	3.74	1200	0.6311	0.8523	0.8490	0.8523	0.8445
0.0809	4.05	1300	0.5321	0.8561	0.8515	0.8561	0.8528
0.0259	4.36	1400	0.8158	0.8408	0.8329	0.8408	0.8288
0.0586	4.67	1500	0.7028	0.8315	0.8359	0.8315	0.8249
0.0218	4.98	1600	0.8059	0.8381	0.8380	0.8381	0.8316
0.0108	5.3	1700	0.7948	0.8474	0.8484	0.8474	0.8470
0.1129	5.61	1800	0.8089	0.8426	0.8492	0.8426	0.8431

Framework versions

Transformers 4.39.0.dev0
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.15.2