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_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: 0.6783333333333333

smids_1x_deit_small_sgd_0001_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: 0.7968
• Accuracy: 0.6783

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
1.0626	1.0	75	1.0608	0.4483
1.0568	2.0	150	1.0464	0.4867
1.0365	3.0	225	1.0338	0.505
1.0078	4.0	300	1.0223	0.5333
0.9982	5.0	375	1.0117	0.5367
0.9993	6.0	450	1.0016	0.5483
0.9996	7.0	525	0.9918	0.5533
0.9666	8.0	600	0.9822	0.5583
0.9644	9.0	675	0.9728	0.5617
0.9545	10.0	750	0.9636	0.575
0.9539	11.0	825	0.9547	0.5817
0.9365	12.0	900	0.9461	0.5867
0.9407	13.0	975	0.9377	0.5983
0.9254	14.0	1050	0.9295	0.6067
0.8947	15.0	1125	0.9216	0.61
0.8953	16.0	1200	0.9139	0.615
0.8981	17.0	1275	0.9064	0.6217
0.8879	18.0	1350	0.8991	0.625
0.8717	19.0	1425	0.8924	0.625
0.894	20.0	1500	0.8856	0.6233
0.8798	21.0	1575	0.8793	0.6267
0.8697	22.0	1650	0.8733	0.6283
0.8459	23.0	1725	0.8674	0.6283
0.8379	24.0	1800	0.8619	0.6317
0.8435	25.0	1875	0.8567	0.63
0.8249	26.0	1950	0.8516	0.6367
0.8188	27.0	2025	0.8468	0.6433
0.8401	28.0	2100	0.8423	0.645
0.8328	29.0	2175	0.8381	0.6483
0.8111	30.0	2250	0.8341	0.66
0.8031	31.0	2325	0.8302	0.6633
0.8167	32.0	2400	0.8267	0.665
0.8141	33.0	2475	0.8233	0.6667
0.7864	34.0	2550	0.8201	0.6717
0.7796	35.0	2625	0.8172	0.6717
0.767	36.0	2700	0.8145	0.675
0.759	37.0	2775	0.8119	0.675
0.7758	38.0	2850	0.8096	0.675
0.7909	39.0	2925	0.8075	0.675
0.7767	40.0	3000	0.8055	0.6767
0.7913	41.0	3075	0.8038	0.6767
0.7892	42.0	3150	0.8023	0.6783
0.787	43.0	3225	0.8010	0.6783
0.7743	44.0	3300	0.7998	0.6783
0.7658	45.0	3375	0.7988	0.6783
0.8086	46.0	3450	0.7980	0.6817
0.7797	47.0	3525	0.7974	0.68
0.7878	48.0	3600	0.7970	0.6783
0.77	49.0	3675	0.7968	0.6783
0.742	50.0	3750	0.7968	0.6783

Framework versions

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