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: hushem_1x_deit_base_rms_00001_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: 0.8048780487804879

hushem_1x_deit_base_rms_00001_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: 0.6686
• Accuracy: 0.8049

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: 1e-05
• 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
No log	1.0	6	1.1610	0.4878
1.1582	2.0	12	0.8046	0.6585
1.1582	3.0	18	0.6473	0.7805
0.4382	4.0	24	0.6375	0.7073
0.1279	5.0	30	0.5568	0.7317
0.1279	6.0	36	0.6762	0.7805
0.0256	7.0	42	0.5322	0.8049
0.0256	8.0	48	0.6244	0.8049
0.0081	9.0	54	0.5937	0.7805
0.0049	10.0	60	0.5932	0.8049
0.0049	11.0	66	0.6107	0.8049
0.0035	12.0	72	0.6136	0.8049
0.0035	13.0	78	0.6118	0.8049
0.0028	14.0	84	0.6111	0.8049
0.0023	15.0	90	0.6113	0.8049
0.0023	16.0	96	0.6216	0.8049
0.0019	17.0	102	0.6208	0.8049
0.0019	18.0	108	0.6301	0.8049
0.0017	19.0	114	0.6331	0.8049
0.0015	20.0	120	0.6358	0.8049
0.0015	21.0	126	0.6382	0.8049
0.0014	22.0	132	0.6393	0.8049
0.0014	23.0	138	0.6427	0.8049
0.0012	24.0	144	0.6480	0.8049
0.0012	25.0	150	0.6496	0.8049
0.0012	26.0	156	0.6549	0.8049
0.0011	27.0	162	0.6526	0.8049
0.0011	28.0	168	0.6563	0.8049
0.001	29.0	174	0.6589	0.8049
0.0009	30.0	180	0.6593	0.8049
0.0009	31.0	186	0.6617	0.8049
0.0009	32.0	192	0.6634	0.8049
0.0009	33.0	198	0.6645	0.8049
0.0009	34.0	204	0.6634	0.8049
0.0008	35.0	210	0.6653	0.8049
0.0008	36.0	216	0.6660	0.8049
0.0008	37.0	222	0.6671	0.8049
0.0008	38.0	228	0.6676	0.8049
0.0008	39.0	234	0.6682	0.8049
0.0008	40.0	240	0.6685	0.8049
0.0008	41.0	246	0.6685	0.8049
0.0008	42.0	252	0.6686	0.8049
0.0008	43.0	258	0.6686	0.8049
0.0007	44.0	264	0.6686	0.8049
0.0008	45.0	270	0.6686	0.8049
0.0008	46.0	276	0.6686	0.8049
0.0008	47.0	282	0.6686	0.8049
0.0008	48.0	288	0.6686	0.8049
0.0008	49.0	294	0.6686	0.8049
0.0008	50.0	300	0.6686	0.8049

Framework versions

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