metadata

license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_10x_deit_tiny_rms_001_fold3
    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: 0.8183333333333334

smids_10x_deit_tiny_rms_001_fold3

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: 0.5645
Accuracy: 0.8183

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: 0.001
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
0.8687	1.0	750	0.8350	0.5417
0.7645	2.0	1500	0.8377	0.545
0.8147	3.0	2250	0.8321	0.5633
0.707	4.0	3000	0.8069	0.5617
0.7456	5.0	3750	0.7498	0.66
0.6732	6.0	4500	0.6947	0.7067
0.7173	7.0	5250	0.6562	0.7233
0.6807	8.0	6000	0.6396	0.735
0.5542	9.0	6750	0.6404	0.725
0.5945	10.0	7500	0.6253	0.715
0.5981	11.0	8250	0.6007	0.7333
0.6124	12.0	9000	0.5926	0.7467
0.5651	13.0	9750	0.6373	0.725
0.5876	14.0	10500	0.6106	0.735
0.595	15.0	11250	0.5814	0.7417
0.6259	16.0	12000	0.6014	0.755
0.5932	17.0	12750	0.6177	0.7433
0.5894	18.0	13500	0.7384	0.68
0.605	19.0	14250	0.6249	0.715
0.5663	20.0	15000	0.6124	0.7367
0.5134	21.0	15750	0.5785	0.7433
0.6186	22.0	16500	0.5747	0.7533
0.5238	23.0	17250	0.5818	0.76
0.5431	24.0	18000	0.5901	0.73
0.5802	25.0	18750	0.5751	0.7583
0.532	26.0	19500	0.6079	0.745
0.4391	27.0	20250	0.5654	0.7683
0.5546	28.0	21000	0.5837	0.7717
0.5308	29.0	21750	0.5546	0.76
0.5138	30.0	22500	0.5584	0.7633
0.4508	31.0	23250	0.5616	0.78
0.4928	32.0	24000	0.5495	0.7683
0.5015	33.0	24750	0.5514	0.7717
0.4951	34.0	25500	0.5352	0.7683
0.47	35.0	26250	0.5246	0.7767
0.4942	36.0	27000	0.5348	0.7833
0.4733	37.0	27750	0.5546	0.7833
0.4787	38.0	28500	0.5356	0.7883
0.4477	39.0	29250	0.5284	0.795
0.5359	40.0	30000	0.5502	0.8
0.4568	41.0	30750	0.5425	0.7883
0.4376	42.0	31500	0.5402	0.79
0.4262	43.0	32250	0.5808	0.7617
0.4405	44.0	33000	0.5553	0.7983
0.3884	45.0	33750	0.5497	0.7783
0.37	46.0	34500	0.5855	0.8067
0.413	47.0	35250	0.5591	0.815
0.3776	48.0	36000	0.5614	0.8067
0.3505	49.0	36750	0.5713	0.805
0.3537	50.0	37500	0.5645	0.8183

Framework versions

Transformers 4.32.1
Pytorch 2.1.1+cu121
Datasets 2.12.0
Tokenizers 0.13.2