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_1x_deit_base_adamax_001_fold1
    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: 0.5333333333333333

hushem_1x_deit_base_adamax_001_fold1

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: 4.0186
• Accuracy: 0.5333

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	1.5701	0.2444
1.7784	2.0	12	1.4475	0.2444
1.7784	3.0	18	1.5245	0.2444
1.3959	4.0	24	1.4708	0.4222
1.2593	5.0	30	1.4970	0.2889
1.2593	6.0	36	2.3855	0.2444
1.1884	7.0	42	1.0899	0.5333
1.1884	8.0	48	1.5303	0.3778
1.0528	9.0	54	1.8879	0.2667
0.9521	10.0	60	1.5051	0.3556
0.9521	11.0	66	1.7383	0.4222
0.832	12.0	72	1.4982	0.4667
0.832	13.0	78	2.2428	0.4222
0.6131	14.0	84	2.0924	0.4667
0.7375	15.0	90	1.9680	0.3778
0.7375	16.0	96	2.1272	0.4
0.3749	17.0	102	2.6064	0.4222
0.3749	18.0	108	3.2946	0.4
0.1952	19.0	114	3.4322	0.4
0.1436	20.0	120	3.6588	0.4222
0.1436	21.0	126	2.5255	0.4889
0.198	22.0	132	4.0901	0.3778
0.198	23.0	138	3.7265	0.4444
0.1826	24.0	144	2.5207	0.5333
0.0757	25.0	150	3.1947	0.4667
0.0757	26.0	156	2.8055	0.5333
0.0579	27.0	162	2.7690	0.5333
0.0579	28.0	168	3.0584	0.5333
0.0219	29.0	174	2.7699	0.6
0.0037	30.0	180	3.6076	0.5556
0.0037	31.0	186	4.1981	0.4667
0.0005	32.0	192	4.2123	0.5111
0.0005	33.0	198	4.1138	0.4889
0.0004	34.0	204	4.0577	0.5333
0.0002	35.0	210	4.0352	0.5333
0.0002	36.0	216	4.0255	0.5333
0.0002	37.0	222	4.0206	0.5333
0.0002	38.0	228	4.0191	0.5333
0.0001	39.0	234	4.0188	0.5333
0.0001	40.0	240	4.0186	0.5333
0.0001	41.0	246	4.0185	0.5333
0.0001	42.0	252	4.0186	0.5333
0.0001	43.0	258	4.0186	0.5333
0.0001	44.0	264	4.0186	0.5333
0.0001	45.0	270	4.0186	0.5333
0.0001	46.0	276	4.0186	0.5333
0.0001	47.0	282	4.0186	0.5333
0.0001	48.0	288	4.0186	0.5333
0.0001	49.0	294	4.0186	0.5333
0.0001	50.0	300	4.0186	0.5333

Framework versions

• Transformers 4.35.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.7
• Tokenizers 0.14.1