metadata

license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_5x_deit_tiny_adamax_00001_fold2
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: test
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8801996672212978

smids_5x_deit_tiny_adamax_00001_fold2

This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.9976
Accuracy: 0.8802

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-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
lr_scheduler_warmup_ratio: 0.1
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.452	1.0	375	0.3866	0.8403
0.2847	2.0	750	0.3266	0.8602
0.2241	3.0	1125	0.3108	0.8602
0.1563	4.0	1500	0.3106	0.8785
0.1317	5.0	1875	0.3206	0.8802
0.0972	6.0	2250	0.3257	0.8835
0.0878	7.0	2625	0.3684	0.8752
0.0825	8.0	3000	0.3750	0.8819
0.0645	9.0	3375	0.4082	0.8852
0.0305	10.0	3750	0.4870	0.8769
0.0215	11.0	4125	0.4928	0.8869
0.037	12.0	4500	0.5391	0.8802
0.0369	13.0	4875	0.6212	0.8719
0.0169	14.0	5250	0.6496	0.8819
0.0205	15.0	5625	0.7009	0.8769
0.0006	16.0	6000	0.7474	0.8735
0.0156	17.0	6375	0.7683	0.8735
0.0004	18.0	6750	0.7918	0.8752
0.0002	19.0	7125	0.8032	0.8819
0.0009	20.0	7500	0.8199	0.8835
0.0001	21.0	7875	0.8709	0.8835
0.0001	22.0	8250	0.8571	0.8785
0.0001	23.0	8625	0.8684	0.8785
0.0001	24.0	9000	0.8915	0.8785
0.0	25.0	9375	0.9054	0.8785
0.0001	26.0	9750	0.9181	0.8802
0.0	27.0	10125	0.9162	0.8785
0.0	28.0	10500	0.9185	0.8802
0.0	29.0	10875	0.9373	0.8769
0.0	30.0	11250	0.9455	0.8819
0.0093	31.0	11625	0.9243	0.8785
0.025	32.0	12000	0.9658	0.8769
0.0144	33.0	12375	0.9598	0.8785
0.0	34.0	12750	0.9760	0.8802
0.0	35.0	13125	0.9707	0.8852
0.0	36.0	13500	0.9857	0.8785
0.0	37.0	13875	0.9774	0.8819
0.0	38.0	14250	0.9769	0.8785
0.0	39.0	14625	0.9854	0.8835
0.0009	40.0	15000	0.9942	0.8769
0.0	41.0	15375	0.9901	0.8802
0.0117	42.0	15750	0.9844	0.8785
0.0061	43.0	16125	0.9978	0.8785
0.0061	44.0	16500	1.0013	0.8802
0.0108	45.0	16875	1.0012	0.8769
0.0	46.0	17250	0.9950	0.8785
0.0106	47.0	17625	0.9952	0.8785
0.0	48.0	18000	0.9951	0.8785
0.0097	49.0	18375	0.9966	0.8785
0.0033	50.0	18750	0.9976	0.8802

Framework versions

Transformers 4.32.1
Pytorch 2.1.1+cu121
Datasets 2.12.0
Tokenizers 0.13.2