metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-eurosat-50
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: Augmented
          split: train
          args: Augmented
        metrics:
          - name: Accuracy
            type: accuracy
            value: 1

swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-eurosat-50

This model is a fine-tuned version of microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.0004
Accuracy: 1.0

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: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.05
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.5952	1.0	55	0.8490	0.6693
0.7582	2.0	110	0.4561	0.8386
0.4359	3.0	165	0.2408	0.9227
0.318	4.0	220	0.1294	0.9568
0.2414	5.0	275	0.0346	0.9909
0.1888	6.0	330	0.0419	0.9864
0.1717	7.0	385	0.0238	0.9943
0.1785	8.0	440	0.0230	0.9943
0.1654	9.0	495	0.0076	1.0
0.1322	10.0	550	0.0046	1.0
0.1123	11.0	605	0.0035	1.0
0.0953	12.0	660	0.0025	1.0
0.0864	13.0	715	0.0033	1.0
0.0984	14.0	770	0.0033	0.9989
0.0952	15.0	825	0.0015	1.0
0.0678	16.0	880	0.0022	1.0
0.0592	17.0	935	0.0013	1.0
0.0729	18.0	990	0.0037	0.9989
0.0672	19.0	1045	0.0041	0.9989
0.0615	20.0	1100	0.0010	1.0
0.058	21.0	1155	0.0009	1.0
0.0571	22.0	1210	0.0021	0.9989
0.0755	23.0	1265	0.0022	0.9989
0.0688	24.0	1320	0.0025	0.9989
0.0417	25.0	1375	0.0003	1.0
0.0589	26.0	1430	0.0007	1.0
0.0563	27.0	1485	0.0007	1.0
0.0603	28.0	1540	0.0010	0.9989
0.0469	29.0	1595	0.0005	1.0
0.0525	30.0	1650	0.0004	1.0

Framework versions

Transformers 4.30.2
Pytorch 2.0.1+cu118
Datasets 2.13.1
Tokenizers 0.13.3