metadata

license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_5x_deit_tiny_sgd_001_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: 0.8818635607321131

smids_5x_deit_tiny_sgd_001_fold2

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: 0.3259
Accuracy: 0.8819

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
0.8524	1.0	375	0.7691	0.6689
0.5007	2.0	750	0.5539	0.7804
0.4114	3.0	1125	0.4742	0.8070
0.3629	4.0	1500	0.4296	0.8286
0.3623	5.0	1875	0.3981	0.8469
0.3098	6.0	2250	0.3783	0.8502
0.3017	7.0	2625	0.3643	0.8453
0.3224	8.0	3000	0.3602	0.8519
0.2666	9.0	3375	0.3471	0.8586
0.2737	10.0	3750	0.3436	0.8552
0.2547	11.0	4125	0.3356	0.8669
0.2986	12.0	4500	0.3379	0.8602
0.2268	13.0	4875	0.3304	0.8669
0.2538	14.0	5250	0.3304	0.8702
0.2279	15.0	5625	0.3282	0.8602
0.1964	16.0	6000	0.3276	0.8719
0.2475	17.0	6375	0.3297	0.8652
0.2224	18.0	6750	0.3277	0.8669
0.1863	19.0	7125	0.3205	0.8686
0.2493	20.0	7500	0.3208	0.8752
0.1873	21.0	7875	0.3214	0.8769
0.1921	22.0	8250	0.3223	0.8735
0.2083	23.0	8625	0.3204	0.8735
0.1865	24.0	9000	0.3201	0.8702
0.1643	25.0	9375	0.3196	0.8802
0.2115	26.0	9750	0.3209	0.8785
0.2108	27.0	10125	0.3192	0.8802
0.1576	28.0	10500	0.3201	0.8802
0.1807	29.0	10875	0.3220	0.8785
0.1891	30.0	11250	0.3216	0.8802
0.1864	31.0	11625	0.3224	0.8835
0.1759	32.0	12000	0.3215	0.8852
0.1618	33.0	12375	0.3224	0.8835
0.1343	34.0	12750	0.3219	0.8835
0.1642	35.0	13125	0.3213	0.8852
0.1538	36.0	13500	0.3239	0.8785
0.1527	37.0	13875	0.3229	0.8852
0.1581	38.0	14250	0.3248	0.8802
0.135	39.0	14625	0.3238	0.8852
0.1591	40.0	15000	0.3237	0.8835
0.1366	41.0	15375	0.3243	0.8819
0.1361	42.0	15750	0.3249	0.8785
0.1751	43.0	16125	0.3245	0.8835
0.135	44.0	16500	0.3255	0.8819
0.1208	45.0	16875	0.3256	0.8819
0.1748	46.0	17250	0.3261	0.8819
0.1449	47.0	17625	0.3264	0.8785
0.1594	48.0	18000	0.3263	0.8785
0.1892	49.0	18375	0.3260	0.8819
0.1218	50.0	18750	0.3259	0.8819

Framework versions

Transformers 4.32.1
Pytorch 2.1.1+cu121
Datasets 2.12.0
Tokenizers 0.13.2