metadata

tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: msi-nat-mini
    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: 0.6308708414872799
          - name: F1
            type: f1
            value: 0.47632740072381147
          - name: Precision
            type: precision
            value: 0.6193914388860238
          - name: Recall
            type: recall
            value: 0.3869512686266613

msi-nat-mini

This model was trained from scratch on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.8600
Accuracy: 0.6309
F1: 0.4763
Precision: 0.6194
Recall: 0.3870

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: 1e-06
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: 0.1
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.5496	1.0	2015	0.7573	0.5955	0.4196	0.5559	0.3369
0.4807	2.0	4031	0.7416	0.6309	0.4981	0.6074	0.4222
0.4235	3.0	6047	0.7680	0.6325	0.5047	0.6076	0.4317
0.3879	4.0	8063	0.7875	0.6339	0.4923	0.6179	0.4092
0.3702	5.0	10078	0.7923	0.6383	0.5128	0.6168	0.4388
0.3568	6.0	12094	0.8311	0.6313	0.4969	0.6090	0.4197
0.3661	7.0	14110	0.8345	0.6316	0.4843	0.6166	0.3987
0.354	8.0	16126	0.8501	0.6305	0.4800	0.6162	0.3931
0.3569	9.0	18141	0.8552	0.6318	0.4809	0.6193	0.3931
0.3536	10.0	20150	0.8600	0.6309	0.4763	0.6194	0.3870

Framework versions

Transformers 4.35.2
Pytorch 2.0.1+cu118
Datasets 2.15.0
Tokenizers 0.15.0