README.md

metadata

license: apache-2.0
base_model: facebook/deit-small-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_1x_deit_small_sgd_0001_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: 0.7054908485856906

smids_1x_deit_small_sgd_0001_fold2

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: 0.7825
• Accuracy: 0.7055

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
1.0783	1.0	75	1.0542	0.4509
1.0523	2.0	150	1.0379	0.4892
1.0294	3.0	225	1.0237	0.5225
1.0365	4.0	300	1.0109	0.5391
1.0245	5.0	375	0.9996	0.5524
0.9804	6.0	450	0.9885	0.5674
1.0004	7.0	525	0.9780	0.5857
0.979	8.0	600	0.9674	0.5957
0.9543	9.0	675	0.9573	0.5973
0.9564	10.0	750	0.9477	0.6023
0.9543	11.0	825	0.9383	0.6190
0.9466	12.0	900	0.9293	0.6256
0.9322	13.0	975	0.9204	0.6273
0.8987	14.0	1050	0.9121	0.6356
0.9075	15.0	1125	0.9040	0.6356
0.9086	16.0	1200	0.8961	0.6456
0.8923	17.0	1275	0.8885	0.6473
0.8806	18.0	1350	0.8814	0.6506
0.8958	19.0	1425	0.8744	0.6522
0.8609	20.0	1500	0.8676	0.6572
0.9085	21.0	1575	0.8614	0.6656
0.8274	22.0	1650	0.8554	0.6689
0.8481	23.0	1725	0.8495	0.6739
0.8248	24.0	1800	0.8441	0.6739
0.8342	25.0	1875	0.8389	0.6722
0.8564	26.0	1950	0.8341	0.6805
0.8458	27.0	2025	0.8294	0.6822
0.7955	28.0	2100	0.8250	0.6839
0.8045	29.0	2175	0.8208	0.6839
0.8063	30.0	2250	0.8168	0.6822
0.8139	31.0	2325	0.8131	0.6855
0.8204	32.0	2400	0.8097	0.6855
0.7688	33.0	2475	0.8065	0.6889
0.8146	34.0	2550	0.8035	0.6938
0.7717	35.0	2625	0.8006	0.6938
0.7969	36.0	2700	0.7981	0.6955
0.805	37.0	2775	0.7957	0.6955
0.8385	38.0	2850	0.7936	0.6988
0.7682	39.0	2925	0.7916	0.6988
0.7759	40.0	3000	0.7898	0.7005
0.8019	41.0	3075	0.7883	0.7005
0.7801	42.0	3150	0.7869	0.7005
0.7773	43.0	3225	0.7857	0.6988
0.788	44.0	3300	0.7847	0.7005
0.7811	45.0	3375	0.7839	0.7022
0.7761	46.0	3450	0.7833	0.7022
0.7855	47.0	3525	0.7828	0.7038
0.7857	48.0	3600	0.7826	0.7055
0.7597	49.0	3675	0.7825	0.7055
0.7828	50.0	3750	0.7825	0.7055

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0