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_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: 0.8935108153078203

smids_10x_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: 0.9769
Accuracy: 0.8935

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
0.2848	1.0	750	0.3201	0.8702
0.1766	2.0	1500	0.2882	0.8869
0.1715	3.0	2250	0.2793	0.8918
0.1528	4.0	3000	0.2928	0.8918
0.1474	5.0	3750	0.3209	0.8902
0.0904	6.0	4500	0.3386	0.8835
0.0506	7.0	5250	0.4033	0.8918
0.0528	8.0	6000	0.4656	0.8985
0.0576	9.0	6750	0.4872	0.8852
0.0247	10.0	7500	0.5636	0.8968
0.0283	11.0	8250	0.5957	0.8918
0.017	12.0	9000	0.6715	0.8952
0.0013	13.0	9750	0.7105	0.8885
0.0004	14.0	10500	0.7432	0.8819
0.0011	15.0	11250	0.7702	0.8968
0.0001	16.0	12000	0.8390	0.8935
0.0001	17.0	12750	0.8598	0.8918
0.0002	18.0	13500	0.8606	0.8852
0.0	19.0	14250	0.8893	0.8902
0.0001	20.0	15000	0.9092	0.8869
0.0	21.0	15750	0.9243	0.8935
0.0414	22.0	16500	0.8856	0.8935
0.0196	23.0	17250	0.9361	0.8902
0.0	24.0	18000	0.9456	0.8952
0.0	25.0	18750	0.9417	0.8935
0.0	26.0	19500	0.9178	0.8968
0.0233	27.0	20250	0.9491	0.8902
0.0	28.0	21000	0.9447	0.9002
0.0	29.0	21750	0.9458	0.8952
0.0	30.0	22500	0.9429	0.8935
0.0	31.0	23250	0.9485	0.8902
0.0	32.0	24000	0.9592	0.8918
0.0	33.0	24750	0.9720	0.8935
0.0	34.0	25500	0.9605	0.8918
0.0	35.0	26250	0.9711	0.8952
0.0	36.0	27000	0.9779	0.8902
0.0248	37.0	27750	0.9824	0.8918
0.0	38.0	28500	0.9776	0.8968
0.0	39.0	29250	0.9729	0.8968
0.0	40.0	30000	0.9687	0.8952
0.0	41.0	30750	0.9781	0.8952
0.0	42.0	31500	0.9860	0.8918
0.0	43.0	32250	0.9836	0.8918
0.0	44.0	33000	0.9765	0.8952
0.0004	45.0	33750	0.9803	0.8935
0.0	46.0	34500	0.9754	0.8952
0.0	47.0	35250	0.9749	0.8952
0.0	48.0	36000	0.9757	0.8952
0.0	49.0	36750	0.9765	0.8935
0.0	50.0	37500	0.9769	0.8935

Framework versions

Transformers 4.32.1
Pytorch 2.1.1+cu121
Datasets 2.12.0
Tokenizers 0.13.2