README.md

metadata

license: apache-2.0
base_model: facebook/deit-small-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_5x_deit_small_adamax_00001_fold1
    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: 0.6888888888888889

hushem_5x_deit_small_adamax_00001_fold1

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: 1.0089
• Accuracy: 0.6889

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
1.2704	1.0	27	1.2632	0.3333
0.9284	2.0	54	1.1387	0.4
0.673	3.0	81	0.9948	0.5111
0.5262	4.0	108	0.8999	0.6222
0.3091	5.0	135	0.8487	0.5778
0.2356	6.0	162	0.7708	0.7333
0.1849	7.0	189	0.7590	0.7111
0.1256	8.0	216	0.7636	0.6889
0.0704	9.0	243	0.7602	0.6444
0.0451	10.0	270	0.7394	0.6667
0.0288	11.0	297	0.7424	0.7111
0.0167	12.0	324	0.7807	0.6667
0.0111	13.0	351	0.8113	0.6667
0.0073	14.0	378	0.8256	0.7111
0.006	15.0	405	0.8473	0.6889
0.0044	16.0	432	0.8545	0.6889
0.0038	17.0	459	0.8649	0.7111
0.0035	18.0	486	0.8829	0.6889
0.0029	19.0	513	0.8931	0.6889
0.0027	20.0	540	0.8979	0.6889
0.0022	21.0	567	0.9159	0.6889
0.0022	22.0	594	0.9078	0.6889
0.002	23.0	621	0.9310	0.6889
0.0018	24.0	648	0.9346	0.6889
0.0018	25.0	675	0.9373	0.6889
0.0017	26.0	702	0.9476	0.6889
0.0016	27.0	729	0.9510	0.6889
0.0014	28.0	756	0.9558	0.6889
0.0015	29.0	783	0.9590	0.6889
0.0013	30.0	810	0.9714	0.6889
0.0012	31.0	837	0.9702	0.6889
0.0012	32.0	864	0.9742	0.6889
0.0011	33.0	891	0.9800	0.6889
0.0011	34.0	918	0.9820	0.6889
0.0011	35.0	945	0.9877	0.6889
0.0011	36.0	972	0.9898	0.6889
0.001	37.0	999	0.9922	0.6889
0.001	38.0	1026	0.9935	0.6889
0.0009	39.0	1053	0.9969	0.6889
0.0009	40.0	1080	0.9993	0.6889
0.0009	41.0	1107	1.0018	0.6889
0.0009	42.0	1134	1.0033	0.6889
0.0009	43.0	1161	1.0054	0.6889
0.0009	44.0	1188	1.0069	0.6889
0.0009	45.0	1215	1.0080	0.6889
0.0009	46.0	1242	1.0085	0.6889
0.0009	47.0	1269	1.0088	0.6889
0.0009	48.0	1296	1.0089	0.6889
0.0009	49.0	1323	1.0089	0.6889
0.0009	50.0	1350	1.0089	0.6889

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0