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_rms_0001_fold4
    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: 0.9761904761904762

hushem_5x_deit_small_rms_0001_fold4

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: 0.2776
• Accuracy: 0.9762

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
1.547	1.0	28	1.3156	0.3571
1.0708	2.0	56	0.7305	0.7381
0.4592	3.0	84	0.4363	0.7857
0.2242	4.0	112	0.3032	0.9048
0.1723	5.0	140	0.4398	0.9048
0.0273	6.0	168	0.1828	0.9286
0.0074	7.0	196	0.2243	0.9286
0.0246	8.0	224	0.6869	0.8095
0.0271	9.0	252	0.5189	0.8810
0.0003	10.0	280	0.4075	0.9048
0.0001	11.0	308	0.3572	0.9286
0.0001	12.0	336	0.3487	0.9286
0.0001	13.0	364	0.3437	0.9286
0.0001	14.0	392	0.3375	0.9286
0.0001	15.0	420	0.3312	0.9286
0.0001	16.0	448	0.3266	0.9286
0.0	17.0	476	0.3232	0.9286
0.0	18.0	504	0.3215	0.9286
0.0	19.0	532	0.3147	0.9286
0.0	20.0	560	0.3112	0.9286
0.0	21.0	588	0.3078	0.9286
0.0	22.0	616	0.3039	0.9524
0.0	23.0	644	0.3017	0.9524
0.0	24.0	672	0.2991	0.9286
0.0	25.0	700	0.2971	0.9524
0.0	26.0	728	0.2950	0.9524
0.0	27.0	756	0.2933	0.9524
0.0	28.0	784	0.2903	0.9524
0.0	29.0	812	0.2892	0.9524
0.0	30.0	840	0.2867	0.9524
0.0	31.0	868	0.2851	0.9524
0.0	32.0	896	0.2853	0.9524
0.0	33.0	924	0.2820	0.9524
0.0	34.0	952	0.2802	0.9524
0.0	35.0	980	0.2774	0.9524
0.0	36.0	1008	0.2771	0.9524
0.0	37.0	1036	0.2772	0.9524
0.0	38.0	1064	0.2772	0.9524
0.0	39.0	1092	0.2769	0.9524
0.0	40.0	1120	0.2766	0.9524
0.0	41.0	1148	0.2767	0.9524
0.0	42.0	1176	0.2776	0.9524
0.0	43.0	1204	0.2777	0.9524
0.0	44.0	1232	0.2779	0.9524
0.0	45.0	1260	0.2777	0.9524
0.0	46.0	1288	0.2776	0.9524
0.0	47.0	1316	0.2776	0.9524
0.0	48.0	1344	0.2776	0.9762
0.0	49.0	1372	0.2776	0.9762
0.0	50.0	1400	0.2776	0.9762

Framework versions

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