metadata

license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_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: 0.5777777777777777

hushem_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: 1.6641
Accuracy: 0.5778

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
1.3635	1.0	27	1.4130	0.2667
1.0042	2.0	54	1.4711	0.2222
0.8161	3.0	81	1.3782	0.2667
0.718	4.0	108	1.4066	0.3778
0.5558	5.0	135	1.3265	0.4444
0.4509	6.0	162	1.2556	0.5111
0.3768	7.0	189	1.2546	0.5333
0.3011	8.0	216	1.2575	0.5333
0.2105	9.0	243	1.2766	0.5111
0.1649	10.0	270	1.2667	0.5333
0.1073	11.0	297	1.2756	0.5333
0.09	12.0	324	1.2325	0.5333
0.0621	13.0	351	1.3118	0.5111
0.0505	14.0	378	1.2588	0.5111
0.0376	15.0	405	1.2895	0.5111
0.0263	16.0	432	1.3784	0.5333
0.0213	17.0	459	1.3797	0.5556
0.0147	18.0	486	1.3696	0.5556
0.0099	19.0	513	1.4119	0.5778
0.0067	20.0	540	1.4307	0.5333
0.0051	21.0	567	1.4626	0.5556
0.0036	22.0	594	1.4677	0.5778
0.0029	23.0	621	1.5080	0.5778
0.0025	24.0	648	1.5082	0.5778
0.002	25.0	675	1.5166	0.5556
0.0019	26.0	702	1.5536	0.5556
0.0018	27.0	729	1.5513	0.5556
0.0016	28.0	756	1.5675	0.5556
0.0015	29.0	783	1.5750	0.5778
0.0014	30.0	810	1.5873	0.5778
0.0012	31.0	837	1.5953	0.6
0.0013	32.0	864	1.6002	0.6
0.001	33.0	891	1.6127	0.6
0.0011	34.0	918	1.6173	0.6
0.0009	35.0	945	1.6219	0.5778
0.001	36.0	972	1.6280	0.6
0.0009	37.0	999	1.6357	0.6
0.0009	38.0	1026	1.6422	0.6
0.0009	39.0	1053	1.6437	0.6
0.0008	40.0	1080	1.6487	0.6
0.0008	41.0	1107	1.6556	0.5778
0.0007	42.0	1134	1.6560	0.6
0.0008	43.0	1161	1.6560	0.6
0.0007	44.0	1188	1.6595	0.6
0.0007	45.0	1215	1.6630	0.5778
0.0007	46.0	1242	1.6626	0.6
0.0007	47.0	1269	1.6639	0.5778
0.0008	48.0	1296	1.6641	0.5778
0.0008	49.0	1323	1.6641	0.5778
0.0007	50.0	1350	1.6641	0.5778

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu118
Datasets 2.15.0
Tokenizers 0.15.0