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_001_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: 0.627906976744186

hushem_1x_deit_tiny_adamax_001_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: 2.8954
• Accuracy: 0.6279

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	1.3703	0.2558
1.9279	2.0	12	1.2966	0.3953
1.9279	3.0	18	1.5490	0.3256
1.3451	4.0	24	1.3082	0.4186
1.2763	5.0	30	1.4000	0.3023
1.2763	6.0	36	1.3783	0.3488
1.1541	7.0	42	1.2878	0.3953
1.1541	8.0	48	1.2528	0.4651
1.0831	9.0	54	1.2761	0.4884
1.0032	10.0	60	0.9439	0.6279
1.0032	11.0	66	1.9597	0.3256
1.0649	12.0	72	1.3501	0.4651
1.0649	13.0	78	1.2845	0.6279
0.8485	14.0	84	1.2102	0.5814
0.7758	15.0	90	1.5993	0.4651
0.7758	16.0	96	1.1744	0.6279
0.5906	17.0	102	1.9493	0.4884
0.5906	18.0	108	1.3370	0.5581
0.5433	19.0	114	1.8704	0.5814
0.4053	20.0	120	2.3449	0.6047
0.4053	21.0	126	2.8071	0.4651
0.6321	22.0	132	1.8750	0.5814
0.6321	23.0	138	1.9591	0.5814
0.2883	24.0	144	2.0517	0.6744
0.2248	25.0	150	2.2716	0.5581
0.2248	26.0	156	2.5758	0.5581
0.0908	27.0	162	2.4971	0.5814
0.0908	28.0	168	2.2990	0.6512
0.0607	29.0	174	2.2806	0.6977
0.0385	30.0	180	2.4187	0.6279
0.0385	31.0	186	2.4113	0.6744
0.0085	32.0	192	2.4630	0.6512
0.0085	33.0	198	2.7214	0.6279
0.004	34.0	204	2.8415	0.6047
0.0007	35.0	210	2.8858	0.6047
0.0007	36.0	216	2.8956	0.6279
0.0005	37.0	222	2.8935	0.6279
0.0005	38.0	228	2.8908	0.6279
0.0004	39.0	234	2.8922	0.6279
0.0003	40.0	240	2.8936	0.6279
0.0003	41.0	246	2.8951	0.6279
0.0003	42.0	252	2.8954	0.6279
0.0003	43.0	258	2.8954	0.6279
0.0003	44.0	264	2.8954	0.6279
0.0003	45.0	270	2.8954	0.6279
0.0003	46.0	276	2.8954	0.6279
0.0003	47.0	282	2.8954	0.6279
0.0003	48.0	288	2.8954	0.6279
0.0003	49.0	294	2.8954	0.6279
0.0003	50.0	300	2.8954	0.6279

Framework versions

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