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: smids_3x_deit_base_rms_0001_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: 0.8933333333333333

smids_3x_deit_base_rms_0001_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: 1.1319
• Accuracy: 0.8933

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
0.3269	1.0	225	0.4237	0.8533
0.1863	2.0	450	0.4104	0.8633
0.1323	3.0	675	0.3455	0.8767
0.0674	4.0	900	0.4806	0.895
0.0524	5.0	1125	0.4637	0.8783
0.0407	6.0	1350	0.4567	0.89
0.0468	7.0	1575	0.5568	0.8767
0.0239	8.0	1800	0.6027	0.8783
0.0176	9.0	2025	0.6627	0.8817
0.0132	10.0	2250	0.7320	0.8683
0.0166	11.0	2475	0.6923	0.88
0.11	12.0	2700	0.5801	0.8883
0.0376	13.0	2925	0.4794	0.89
0.0285	14.0	3150	0.6473	0.8883
0.0192	15.0	3375	0.7068	0.8967
0.0041	16.0	3600	0.7011	0.895
0.012	17.0	3825	0.6525	0.9017
0.03	18.0	4050	0.6508	0.91
0.0251	19.0	4275	0.7493	0.8967
0.0108	20.0	4500	0.7077	0.895
0.0009	21.0	4725	0.6790	0.89
0.0002	22.0	4950	0.7411	0.8967
0.0264	23.0	5175	0.7794	0.8983
0.0051	24.0	5400	0.9553	0.8883
0.0221	25.0	5625	0.7771	0.905
0.0315	26.0	5850	0.7638	0.9
0.003	27.0	6075	0.8047	0.9
0.0125	28.0	6300	0.7560	0.9
0.0039	29.0	6525	0.7149	0.9067
0.0	30.0	6750	0.8257	0.9
0.0	31.0	6975	0.8249	0.9133
0.0	32.0	7200	0.8553	0.9033
0.01	33.0	7425	0.9333	0.895
0.0	34.0	7650	0.9286	0.9067
0.0024	35.0	7875	0.9413	0.8983
0.0	36.0	8100	0.8868	0.9083
0.0039	37.0	8325	0.9484	0.9033
0.0	38.0	8550	0.9617	0.9033
0.0	39.0	8775	0.9572	0.9017
0.0	40.0	9000	1.0465	0.8933
0.0	41.0	9225	1.0197	0.8983
0.0	42.0	9450	1.0477	0.895
0.0029	43.0	9675	1.0659	0.8983
0.0	44.0	9900	1.0846	0.8967
0.0	45.0	10125	1.1008	0.8983
0.0	46.0	10350	1.1123	0.8917
0.0	47.0	10575	1.1192	0.8933
0.0	48.0	10800	1.1251	0.8933
0.0	49.0	11025	1.1289	0.8933
0.0	50.0	11250	1.1319	0.8933

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2