metadata

license: apache-2.0
base_model: facebook/deit-small-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_3x_deit_small_rms_001_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: 0.7033333333333334

smids_3x_deit_small_rms_001_fold5

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

Loss: 0.6403
Accuracy: 0.7033

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
1.1309	1.0	225	1.0865	0.4083
0.9745	2.0	450	0.8750	0.56
0.8495	3.0	675	0.8352	0.5333
0.841	4.0	900	0.7966	0.555
0.8226	5.0	1125	0.7921	0.6
0.853	6.0	1350	0.8095	0.5533
0.8653	7.0	1575	0.9709	0.5433
0.819	8.0	1800	0.7655	0.6383
0.785	9.0	2025	0.7896	0.625
0.8631	10.0	2250	0.8879	0.525
0.8663	11.0	2475	0.8802	0.535
0.7946	12.0	2700	0.8056	0.6
0.881	13.0	2925	0.8010	0.5733
0.8182	14.0	3150	0.7603	0.6483
0.7224	15.0	3375	0.7710	0.645
0.7369	16.0	3600	0.7559	0.615
0.7647	17.0	3825	0.7835	0.5983
0.7746	18.0	4050	0.7447	0.645
0.7626	19.0	4275	0.7290	0.66
0.7327	20.0	4500	0.7697	0.6333
0.7163	21.0	4725	0.7314	0.6433
0.8028	22.0	4950	0.7791	0.6
0.7667	23.0	5175	0.7694	0.64
0.7558	24.0	5400	0.7110	0.6567
0.7363	25.0	5625	0.7150	0.655
0.7997	26.0	5850	0.7280	0.6617
0.7362	27.0	6075	0.7643	0.625
0.7462	28.0	6300	0.7414	0.6233
0.7001	29.0	6525	0.6980	0.665
0.6874	30.0	6750	0.7253	0.6283
0.6271	31.0	6975	0.7291	0.6433
0.6551	32.0	7200	0.6785	0.6567
0.7073	33.0	7425	0.6937	0.6717
0.6958	34.0	7650	0.7523	0.6367
0.6565	35.0	7875	0.6954	0.6533
0.6367	36.0	8100	0.6638	0.6817
0.6915	37.0	8325	0.6412	0.7083
0.5928	38.0	8550	0.7119	0.6533
0.6501	39.0	8775	0.6232	0.7133
0.5902	40.0	9000	0.6501	0.71
0.6466	41.0	9225	0.6743	0.6683
0.7124	42.0	9450	0.7091	0.6783
0.6189	43.0	9675	0.6361	0.71
0.6514	44.0	9900	0.6443	0.7
0.6016	45.0	10125	0.6523	0.6883
0.5823	46.0	10350	0.6473	0.685
0.6451	47.0	10575	0.6348	0.69
0.5906	48.0	10800	0.6387	0.695
0.5662	49.0	11025	0.6330	0.7017
0.6218	50.0	11250	0.6403	0.7033

Framework versions

Transformers 4.32.1
Pytorch 2.1.0+cu121
Datasets 2.12.0
Tokenizers 0.13.2