README.md

metadata

license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_1x_deit_tiny_rms_lr0001_fold4
    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: 0.7619047619047619

hushem_1x_deit_tiny_rms_lr0001_fold4

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

• Loss: 1.2563
• Accuracy: 0.7619

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	1.5375	0.2619
2.1724	2.0	12	2.0732	0.2381
2.1724	3.0	18	1.4131	0.2619
1.6782	4.0	24	1.3425	0.4524
1.4417	5.0	30	1.3458	0.2857
1.4417	6.0	36	1.2594	0.6429
1.3676	7.0	42	1.1740	0.4762
1.3676	8.0	48	1.2511	0.3571
1.2512	9.0	54	0.8438	0.6190
0.8279	10.0	60	1.0096	0.5
0.8279	11.0	66	0.7631	0.6667
0.5322	12.0	72	0.6526	0.7857
0.5322	13.0	78	0.6963	0.7143
0.257	14.0	84	0.7429	0.7619
0.1198	15.0	90	0.9632	0.6905
0.1198	16.0	96	1.2325	0.7143
0.0178	17.0	102	1.2090	0.7381
0.0178	18.0	108	1.1054	0.7619
0.0016	19.0	114	1.2184	0.7143
0.0009	20.0	120	1.1716	0.7619
0.0009	21.0	126	1.1784	0.7619
0.0004	22.0	132	1.1866	0.7619
0.0004	23.0	138	1.1935	0.7619
0.0003	24.0	144	1.1995	0.7619
0.0003	25.0	150	1.2046	0.7619
0.0003	26.0	156	1.2111	0.7619
0.0003	27.0	162	1.2169	0.7619
0.0003	28.0	168	1.2218	0.7619
0.0002	29.0	174	1.2261	0.7619
0.0002	30.0	180	1.2318	0.7619
0.0002	31.0	186	1.2354	0.7619
0.0002	32.0	192	1.2392	0.7619
0.0002	33.0	198	1.2423	0.7619
0.0002	34.0	204	1.2453	0.7619
0.0002	35.0	210	1.2477	0.7619
0.0002	36.0	216	1.2499	0.7619
0.0002	37.0	222	1.2519	0.7619
0.0002	38.0	228	1.2534	0.7619
0.0002	39.0	234	1.2547	0.7619
0.0002	40.0	240	1.2556	0.7619
0.0002	41.0	246	1.2562	0.7619
0.0002	42.0	252	1.2563	0.7619
0.0002	43.0	258	1.2563	0.7619
0.0002	44.0	264	1.2563	0.7619
0.0002	45.0	270	1.2563	0.7619
0.0002	46.0	276	1.2563	0.7619
0.0002	47.0	282	1.2563	0.7619
0.0002	48.0	288	1.2563	0.7619
0.0002	49.0	294	1.2563	0.7619
0.0002	50.0	300	1.2563	0.7619

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1