README.md

metadata

license: apache-2.0
base_model: microsoft/beit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_1x_beit_base_adamax_0001_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: 0.8604651162790697

hushem_1x_beit_base_adamax_0001_fold3

This model is a fine-tuned version of microsoft/beit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.5855
• Accuracy: 0.8605

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	1.0849	0.5349
1.2852	2.0	12	0.7460	0.7907
1.2852	3.0	18	0.5699	0.8140
0.4305	4.0	24	0.3649	0.8605
0.1805	5.0	30	0.2406	0.9535
0.1805	6.0	36	0.4656	0.8837
0.0211	7.0	42	0.4915	0.8605
0.0211	8.0	48	0.5042	0.8372
0.0066	9.0	54	0.6760	0.7907
0.0025	10.0	60	0.6098	0.8605
0.0025	11.0	66	0.6353	0.9070
0.0011	12.0	72	0.6882	0.8837
0.0011	13.0	78	0.6437	0.8837
0.0022	14.0	84	0.5430	0.8605
0.0007	15.0	90	0.5436	0.8605
0.0007	16.0	96	0.5847	0.8605
0.0007	17.0	102	0.7054	0.8605
0.0007	18.0	108	0.7624	0.8372
0.0006	19.0	114	0.6619	0.8605
0.0007	20.0	120	0.6238	0.8372
0.0007	21.0	126	0.6086	0.8372
0.0003	22.0	132	0.6074	0.8372
0.0003	23.0	138	0.6228	0.8605
0.0003	24.0	144	0.6265	0.8605
0.0003	25.0	150	0.6139	0.8372
0.0003	26.0	156	0.6063	0.8372
0.0002	27.0	162	0.5981	0.8372
0.0002	28.0	168	0.5901	0.8372
0.0002	29.0	174	0.5785	0.8605
0.0001	30.0	180	0.5753	0.8605
0.0001	31.0	186	0.5775	0.8605
0.0002	32.0	192	0.5781	0.8605
0.0002	33.0	198	0.5782	0.8605
0.0002	34.0	204	0.5804	0.8605
0.0003	35.0	210	0.5817	0.8605
0.0003	36.0	216	0.5823	0.8605
0.0001	37.0	222	0.5831	0.8605
0.0001	38.0	228	0.5855	0.8605
0.0002	39.0	234	0.5859	0.8605
0.0002	40.0	240	0.5858	0.8605
0.0002	41.0	246	0.5855	0.8605
0.0002	42.0	252	0.5855	0.8605
0.0002	43.0	258	0.5855	0.8605
0.0002	44.0	264	0.5855	0.8605
0.0001	45.0	270	0.5855	0.8605
0.0001	46.0	276	0.5855	0.8605
0.0005	47.0	282	0.5855	0.8605
0.0005	48.0	288	0.5855	0.8605
0.0002	49.0	294	0.5855	0.8605
0.0001	50.0	300	0.5855	0.8605

Framework versions

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