README.md

metadata

license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_1x_deit_tiny_adamax_00001_fold5
    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: 0.5121951219512195

hushem_1x_deit_tiny_adamax_00001_fold5

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: 0.9982
• Accuracy: 0.5122

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
No log	1.0	6	1.3816	0.2927
1.4331	2.0	12	1.3595	0.2195
1.4331	3.0	18	1.3006	0.2927
1.2071	4.0	24	1.2477	0.3415
1.0931	5.0	30	1.2218	0.3659
1.0931	6.0	36	1.1904	0.3415
0.9583	7.0	42	1.2070	0.3659
0.9583	8.0	48	1.1804	0.3415
0.875	9.0	54	1.1663	0.3415
0.7821	10.0	60	1.1729	0.3659
0.7821	11.0	66	1.1600	0.3659
0.7082	12.0	72	1.1535	0.3659
0.7082	13.0	78	1.1283	0.3902
0.5865	14.0	84	1.1050	0.4146
0.5549	15.0	90	1.0989	0.4146
0.5549	16.0	96	1.0902	0.4146
0.4748	17.0	102	1.0889	0.4146
0.4748	18.0	108	1.0670	0.4146
0.4005	19.0	114	1.0529	0.4146
0.3717	20.0	120	1.0514	0.4146
0.3717	21.0	126	1.0589	0.4146
0.3189	22.0	132	1.0546	0.4146
0.3189	23.0	138	1.0253	0.4390
0.2768	24.0	144	1.0205	0.4390
0.2632	25.0	150	1.0386	0.4146
0.2632	26.0	156	1.0297	0.4390
0.2284	27.0	162	1.0322	0.4634
0.2284	28.0	168	1.0102	0.4634
0.196	29.0	174	1.0015	0.4878
0.1861	30.0	180	1.0070	0.4634
0.1861	31.0	186	1.0149	0.4878
0.1711	32.0	192	1.0173	0.4878
0.1711	33.0	198	1.0083	0.4878
0.1508	34.0	204	1.0068	0.5122
0.1433	35.0	210	0.9998	0.5122
0.1433	36.0	216	0.9984	0.5122
0.1371	37.0	222	0.9985	0.5122
0.1371	38.0	228	0.9983	0.5122
0.1311	39.0	234	0.9983	0.5122
0.1245	40.0	240	0.9977	0.5122
0.1245	41.0	246	0.9980	0.5122
0.1273	42.0	252	0.9982	0.5122
0.1273	43.0	258	0.9982	0.5122
0.1185	44.0	264	0.9982	0.5122
0.1259	45.0	270	0.9982	0.5122
0.1259	46.0	276	0.9982	0.5122
0.1239	47.0	282	0.9982	0.5122
0.1239	48.0	288	0.9982	0.5122
0.1264	49.0	294	0.9982	0.5122
0.1234	50.0	300	0.9982	0.5122

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1