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_1x_deit_small_adamax_001_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: 0.5853658536585366

hushem_1x_deit_small_adamax_001_fold5

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: 2.9089
• Accuracy: 0.5854

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: 0.001
• 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	1.5709	0.2683
1.6365	2.0	12	1.3231	0.3171
1.6365	3.0	18	1.0858	0.4878
1.2777	4.0	24	1.0527	0.4634
1.0819	5.0	30	2.4025	0.4878
1.0819	6.0	36	1.0776	0.6098
1.1957	7.0	42	1.2491	0.4878
1.1957	8.0	48	1.2390	0.4878
1.0582	9.0	54	2.4696	0.3659
0.9645	10.0	60	0.9800	0.6585
0.9645	11.0	66	1.4465	0.4878
0.7158	12.0	72	1.3709	0.4146
0.7158	13.0	78	1.8787	0.5610
0.4707	14.0	84	2.2003	0.4878
0.3746	15.0	90	2.7652	0.4390
0.3746	16.0	96	1.4738	0.6098
0.3815	17.0	102	2.1297	0.4878
0.3815	18.0	108	2.7358	0.4634
0.2562	19.0	114	2.1602	0.6341
0.215	20.0	120	2.4495	0.5122
0.215	21.0	126	2.2161	0.5366
0.0855	22.0	132	2.6756	0.5610
0.0855	23.0	138	3.4355	0.5366
0.0976	24.0	144	2.8453	0.6098
0.0588	25.0	150	2.9043	0.5854
0.0588	26.0	156	3.0589	0.4878
0.0051	27.0	162	2.7256	0.6098
0.0051	28.0	168	2.6655	0.6098
0.0018	29.0	174	2.6795	0.6098
0.005	30.0	180	2.7568	0.6098
0.005	31.0	186	2.8042	0.6098
0.0004	32.0	192	2.8224	0.6098
0.0004	33.0	198	2.8428	0.6098
0.0002	34.0	204	2.8628	0.5854
0.0002	35.0	210	2.8783	0.5854
0.0002	36.0	216	2.8881	0.5854
0.0002	37.0	222	2.8950	0.5854
0.0002	38.0	228	2.9002	0.5854
0.0002	39.0	234	2.9045	0.5854
0.0002	40.0	240	2.9071	0.5854
0.0002	41.0	246	2.9084	0.5854
0.0001	42.0	252	2.9089	0.5854
0.0001	43.0	258	2.9089	0.5854
0.0002	44.0	264	2.9089	0.5854
0.0001	45.0	270	2.9089	0.5854
0.0001	46.0	276	2.9089	0.5854
0.0002	47.0	282	2.9089	0.5854
0.0002	48.0	288	2.9089	0.5854
0.0001	49.0	294	2.9089	0.5854
0.0001	50.0	300	2.9089	0.5854

Framework versions

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