README.md

metadata

license: apache-2.0
base_model: facebook/deit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_5x_deit_base_adamax_00001_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: 0.8372093023255814

hushem_5x_deit_base_adamax_00001_fold3

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: 0.4419
• Accuracy: 0.8372

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: 1e-05
• 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
1.3275	1.0	28	1.2372	0.5814
1.0641	2.0	56	1.0484	0.6977
0.7591	3.0	84	0.8760	0.7442
0.5652	4.0	112	0.7360	0.8140
0.3906	5.0	140	0.6489	0.8372
0.3059	6.0	168	0.5954	0.8605
0.1994	7.0	196	0.5269	0.8372
0.134	8.0	224	0.5174	0.8605
0.0783	9.0	252	0.4602	0.8605
0.0454	10.0	280	0.4569	0.8372
0.0318	11.0	308	0.4393	0.8837
0.018	12.0	336	0.4222	0.8605
0.0132	13.0	364	0.4453	0.8837
0.0088	14.0	392	0.4098	0.8837
0.0068	15.0	420	0.4226	0.8605
0.0058	16.0	448	0.4268	0.8605
0.0055	17.0	476	0.4132	0.8605
0.0045	18.0	504	0.4342	0.8605
0.004	19.0	532	0.4228	0.8605
0.0033	20.0	560	0.4271	0.8372
0.0033	21.0	588	0.4254	0.8372
0.0029	22.0	616	0.4205	0.8372
0.0027	23.0	644	0.4207	0.8372
0.0024	24.0	672	0.4248	0.8605
0.0022	25.0	700	0.4229	0.8372
0.0021	26.0	728	0.4293	0.8372
0.002	27.0	756	0.4267	0.8372
0.002	28.0	784	0.4239	0.8605
0.0018	29.0	812	0.4273	0.8372
0.0018	30.0	840	0.4313	0.8372
0.0016	31.0	868	0.4289	0.8372
0.0016	32.0	896	0.4329	0.8372
0.0016	33.0	924	0.4313	0.8372
0.0014	34.0	952	0.4362	0.8372
0.0016	35.0	980	0.4336	0.8372
0.0014	36.0	1008	0.4353	0.8372
0.0014	37.0	1036	0.4446	0.8372
0.0013	38.0	1064	0.4482	0.8372
0.0013	39.0	1092	0.4496	0.8372
0.0012	40.0	1120	0.4442	0.8372
0.0013	41.0	1148	0.4456	0.8372
0.0013	42.0	1176	0.4450	0.8372
0.0012	43.0	1204	0.4433	0.8372
0.0012	44.0	1232	0.4424	0.8372
0.0011	45.0	1260	0.4418	0.8372
0.0011	46.0	1288	0.4417	0.8372
0.0011	47.0	1316	0.4421	0.8372
0.0011	48.0	1344	0.4419	0.8372
0.0011	49.0	1372	0.4419	0.8372
0.0011	50.0	1400	0.4419	0.8372

Framework versions

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