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_rms_lr0001_fold3
    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: 0.46511627906976744

hushem_1x_deit_rms_lr0001_fold3

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: 2.7920
• Accuracy: 0.4651

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
No log	1.0	6	1.3959	0.2558
2.0191	2.0	12	1.4540	0.2791
2.0191	3.0	18	1.5040	0.3721
1.4688	4.0	24	1.3687	0.3256
1.3397	5.0	30	1.3082	0.4186
1.3397	6.0	36	1.3917	0.3256
1.1986	7.0	42	1.4209	0.3256
1.1986	8.0	48	1.4510	0.3721
1.0698	9.0	54	1.4225	0.3023
0.8214	10.0	60	1.5289	0.4186
0.8214	11.0	66	1.4884	0.4419
0.5823	12.0	72	2.0101	0.3256
0.5823	13.0	78	1.6036	0.5349
0.4001	14.0	84	1.6332	0.4186
0.2362	15.0	90	2.0095	0.4884
0.2362	16.0	96	1.8563	0.5581
0.1078	17.0	102	2.1555	0.5116
0.1078	18.0	108	2.0019	0.5581
0.0769	19.0	114	2.3852	0.4884
0.0351	20.0	120	2.4880	0.5349
0.0351	21.0	126	2.5950	0.4884
0.001	22.0	132	2.5992	0.4884
0.001	23.0	138	2.6117	0.4884
0.0006	24.0	144	2.6223	0.4884
0.0005	25.0	150	2.6443	0.4884
0.0005	26.0	156	2.6672	0.4884
0.0004	27.0	162	2.6883	0.4884
0.0004	28.0	168	2.6994	0.4884
0.0003	29.0	174	2.7093	0.4884
0.0003	30.0	180	2.7225	0.4884
0.0003	31.0	186	2.7350	0.4884
0.0003	32.0	192	2.7468	0.4651
0.0003	33.0	198	2.7564	0.4651
0.0003	34.0	204	2.7644	0.4651
0.0002	35.0	210	2.7717	0.4651
0.0002	36.0	216	2.7756	0.4651
0.0002	37.0	222	2.7805	0.4651
0.0002	38.0	228	2.7848	0.4651
0.0002	39.0	234	2.7876	0.4651
0.0002	40.0	240	2.7903	0.4651
0.0002	41.0	246	2.7917	0.4651
0.0002	42.0	252	2.7920	0.4651
0.0002	43.0	258	2.7920	0.4651
0.0002	44.0	264	2.7920	0.4651
0.0002	45.0	270	2.7920	0.4651
0.0002	46.0	276	2.7920	0.4651
0.0002	47.0	282	2.7920	0.4651
0.0002	48.0	288	2.7920	0.4651
0.0002	49.0	294	2.7920	0.4651
0.0002	50.0	300	2.7920	0.4651

Framework versions

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