metadata

license: apache-2.0
base_model: facebook/deit-small-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_5x_deit_small_sgd_0001_fold5
    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: 0.8116666666666666

smids_5x_deit_small_sgd_0001_fold5

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

Loss: 0.4899
Accuracy: 0.8117

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: 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
1.0575	1.0	375	1.0409	0.4667
0.9896	2.0	750	1.0031	0.5117
0.9428	3.0	1125	0.9645	0.5567
0.9186	4.0	1500	0.9265	0.615
0.8922	5.0	1875	0.8895	0.6483
0.8541	6.0	2250	0.8539	0.6717
0.7885	7.0	2625	0.8194	0.69
0.7714	8.0	3000	0.7879	0.705
0.758	9.0	3375	0.7592	0.7133
0.7212	10.0	3750	0.7334	0.7217
0.6793	11.0	4125	0.7102	0.7333
0.6484	12.0	4500	0.6895	0.7367
0.6765	13.0	4875	0.6713	0.7467
0.664	14.0	5250	0.6548	0.7533
0.6332	15.0	5625	0.6395	0.7617
0.5983	16.0	6000	0.6261	0.77
0.6122	17.0	6375	0.6142	0.77
0.5912	18.0	6750	0.6024	0.7733
0.5764	19.0	7125	0.5918	0.775
0.5461	20.0	7500	0.5824	0.7783
0.5245	21.0	7875	0.5733	0.7833
0.5339	22.0	8250	0.5654	0.7867
0.5651	23.0	8625	0.5584	0.7867
0.5365	24.0	9000	0.5518	0.7933
0.4982	25.0	9375	0.5457	0.795
0.5274	26.0	9750	0.5402	0.7933
0.5167	27.0	10125	0.5353	0.795
0.53	28.0	10500	0.5303	0.7967
0.5404	29.0	10875	0.5260	0.7967
0.4414	30.0	11250	0.5222	0.8017
0.5269	31.0	11625	0.5183	0.8017
0.5299	32.0	12000	0.5150	0.8017
0.5311	33.0	12375	0.5120	0.8033
0.499	34.0	12750	0.5091	0.8033
0.4712	35.0	13125	0.5065	0.8033
0.4169	36.0	13500	0.5042	0.8017
0.4803	37.0	13875	0.5020	0.8017
0.4796	38.0	14250	0.5001	0.805
0.4865	39.0	14625	0.4984	0.8067
0.5122	40.0	15000	0.4967	0.8083
0.4785	41.0	15375	0.4953	0.8067
0.4562	42.0	15750	0.4941	0.8083
0.5248	43.0	16125	0.4930	0.8117
0.4817	44.0	16500	0.4922	0.8117
0.4662	45.0	16875	0.4914	0.8117
0.4968	46.0	17250	0.4908	0.8117
0.5157	47.0	17625	0.4904	0.8117
0.4378	48.0	18000	0.4901	0.8117
0.4668	49.0	18375	0.4899	0.8117
0.4722	50.0	18750	0.4899	0.8117

Framework versions

Transformers 4.32.1
Pytorch 2.1.1+cu121
Datasets 2.12.0
Tokenizers 0.13.2