scene_classification

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

• Loss: 0.6106
• Accuracy: 0.8492

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.3172	1.0	341	2.8572	0.5109
2.2254	2.0	682	2.1453	0.6329
1.6202	3.0	1023	1.6283	0.7336
1.2313	4.0	1364	1.3402	0.7599
0.9576	5.0	1705	1.1237	0.8010
0.7654	6.0	2046	1.0270	0.8023
0.6416	7.0	2387	0.8848	0.8171
0.5353	8.0	2728	0.8381	0.8087
0.4516	9.0	3069	0.7570	0.8254
0.3925	10.0	3410	0.6667	0.8524
0.3453	11.0	3751	0.7583	0.8164
0.2944	12.0	4092	0.6783	0.8350
0.294	13.0	4433	0.7128	0.8312
0.2507	14.0	4774	0.6632	0.8331
0.2355	15.0	5115	0.6730	0.8421
0.2267	16.0	5456	0.6572	0.8357
0.2032	17.0	5797	0.7058	0.8280
0.1908	18.0	6138	0.6374	0.8485
0.1857	19.0	6479	0.6831	0.8312
0.1727	20.0	6820	0.6961	0.8254
0.1692	21.0	7161	0.6306	0.8402
0.1642	22.0	7502	0.6291	0.8485
0.1618	23.0	7843	0.6058	0.8582
0.1593	24.0	8184	0.6780	0.8389
0.1399	25.0	8525	0.6330	0.8485
0.1373	26.0	8866	0.6550	0.8408
0.1334	27.0	9207	0.6857	0.8421
0.1388	28.0	9548	0.6338	0.8415
0.1423	29.0	9889	0.6272	0.8517
0.1288	30.0	10230	0.6409	0.8556

Framework versions

• Transformers 4.33.3
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3