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_fold5
    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: 0.8780487804878049

hushem_5x_deit_base_adamax_00001_fold5

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: 0.5126
Accuracy: 0.8780

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
1.2842	1.0	28	1.2327	0.4634
1.0428	2.0	56	1.0859	0.5854
0.7338	3.0	84	0.9001	0.6585
0.526	4.0	112	0.7803	0.6829
0.3644	5.0	140	0.6952	0.6829
0.2797	6.0	168	0.5854	0.7073
0.1574	7.0	196	0.5546	0.7561
0.1102	8.0	224	0.5073	0.7805
0.0668	9.0	252	0.4703	0.8049
0.0352	10.0	280	0.4612	0.8293
0.0202	11.0	308	0.4468	0.8537
0.0126	12.0	336	0.4508	0.8537
0.0083	13.0	364	0.4495	0.8537
0.0078	14.0	392	0.4584	0.8780
0.0058	15.0	420	0.4565	0.8537
0.0051	16.0	448	0.4617	0.8780
0.0041	17.0	476	0.4620	0.8537
0.0038	18.0	504	0.4684	0.8780
0.0035	19.0	532	0.4714	0.8780
0.0033	20.0	560	0.4764	0.8780
0.0027	21.0	588	0.4824	0.8780
0.0027	22.0	616	0.4828	0.8780
0.0023	23.0	644	0.4858	0.8780
0.0021	24.0	672	0.4874	0.8780
0.002	25.0	700	0.4903	0.8780
0.002	26.0	728	0.4912	0.8780
0.0019	27.0	756	0.4926	0.8780
0.0017	28.0	784	0.4940	0.8780
0.0016	29.0	812	0.4962	0.8780
0.0016	30.0	840	0.4964	0.8780
0.0015	31.0	868	0.4991	0.8780
0.0015	32.0	896	0.5002	0.8780
0.0013	33.0	924	0.5024	0.8780
0.0014	34.0	952	0.5035	0.8780
0.0013	35.0	980	0.5045	0.8780
0.0013	36.0	1008	0.5049	0.8780
0.0012	37.0	1036	0.5067	0.8780
0.0012	38.0	1064	0.5085	0.8780
0.0012	39.0	1092	0.5083	0.8780
0.0011	40.0	1120	0.5095	0.8780
0.0011	41.0	1148	0.5101	0.8780
0.0011	42.0	1176	0.5103	0.8780
0.0011	43.0	1204	0.5116	0.8780
0.0011	44.0	1232	0.5122	0.8780
0.0011	45.0	1260	0.5120	0.8780
0.0011	46.0	1288	0.5123	0.8780
0.001	47.0	1316	0.5124	0.8780
0.0011	48.0	1344	0.5126	0.8780
0.001	49.0	1372	0.5126	0.8780
0.0011	50.0	1400	0.5126	0.8780

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu118
Datasets 2.15.0
Tokenizers 0.15.0