metadata

license: apache-2.0
base_model: facebook/deit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_5x_deit_base_adamax_0001_fold2
    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: 0.8985024958402662

smids_5x_deit_base_adamax_0001_fold2

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

Loss: 0.8969
Accuracy: 0.8985

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
0.2012	1.0	375	0.2802	0.8835
0.1424	2.0	750	0.3530	0.8985
0.0484	3.0	1125	0.5841	0.8769
0.0171	4.0	1500	0.5393	0.8968
0.0016	5.0	1875	0.6879	0.8835
0.0133	6.0	2250	0.7421	0.8885
0.0004	7.0	2625	0.7382	0.8869
0.0224	8.0	3000	0.6881	0.8902
0.0004	9.0	3375	0.7760	0.8902
0.0002	10.0	3750	0.7986	0.8852
0.0045	11.0	4125	0.7173	0.8935
0.0002	12.0	4500	0.8875	0.8802
0.0106	13.0	4875	0.8591	0.8918
0.0009	14.0	5250	0.9035	0.8902
0.0101	15.0	5625	0.8626	0.8918
0.0	16.0	6000	0.9182	0.8852
0.0029	17.0	6375	0.7794	0.8952
0.0	18.0	6750	0.7848	0.8935
0.0001	19.0	7125	0.8673	0.8902
0.0	20.0	7500	0.8428	0.8918
0.0	21.0	7875	0.8282	0.8952
0.0	22.0	8250	0.8604	0.8918
0.0	23.0	8625	0.8223	0.8935
0.0	24.0	9000	0.8436	0.8952
0.0	25.0	9375	0.8078	0.8902
0.0	26.0	9750	0.8487	0.8968
0.0	27.0	10125	0.8273	0.8902
0.0	28.0	10500	0.8385	0.8902
0.0	29.0	10875	0.8210	0.8985
0.0	30.0	11250	0.8440	0.8918
0.0029	31.0	11625	0.8614	0.8852
0.0034	32.0	12000	0.8524	0.8935
0.0033	33.0	12375	0.8611	0.8918
0.0	34.0	12750	0.8778	0.8985
0.0	35.0	13125	0.8525	0.8952
0.0	36.0	13500	0.8763	0.8952
0.0	37.0	13875	0.8733	0.9002
0.0	38.0	14250	0.8847	0.8952
0.0	39.0	14625	0.8741	0.8952
0.0027	40.0	15000	0.8864	0.8952
0.0	41.0	15375	0.8807	0.8952
0.0025	42.0	15750	0.8886	0.8952
0.0024	43.0	16125	0.8857	0.8985
0.0024	44.0	16500	0.8867	0.8968
0.0023	45.0	16875	0.8921	0.8985
0.0	46.0	17250	0.8968	0.8985
0.0048	47.0	17625	0.8952	0.8985
0.0	48.0	18000	0.8977	0.8985
0.0023	49.0	18375	0.8974	0.8985
0.0023	50.0	18750	0.8969	0.8985

Framework versions

Transformers 4.32.1
Pytorch 2.1.0+cu121
Datasets 2.12.0
Tokenizers 0.13.2