metadata

license: apache-2.0
base_model: microsoft/swinv2-tiny-patch4-window8-256
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: swinv2-tiny-patch4-window8-256-DMAE-ex
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: validation
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.45652173913043476

swinv2-tiny-patch4-window8-256-DMAE-ex

This model is a fine-tuned version of microsoft/swinv2-tiny-patch4-window8-256 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 1.2148
Accuracy: 0.4565

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: 0.004
train_batch_size: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.86	3	6.6832	0.1087
No log	2.0	7	1.2148	0.4565
4.4686	2.86	10	2.5061	0.3261
4.4686	4.0	14	1.4142	0.4565
4.4686	4.86	17	1.6118	0.4565
1.7414	6.0	21	1.2484	0.4565
1.7414	6.86	24	1.3690	0.3261
1.7414	8.0	28	1.4065	0.4565
1.3568	8.86	31	1.2682	0.3261
1.3568	10.0	35	1.2140	0.4565
1.3568	10.86	38	1.2591	0.4565
1.2275	12.0	42	1.2519	0.4565
1.2275	12.86	45	1.2184	0.4565
1.2275	14.0	49	1.2592	0.4565
1.3025	14.86	52	1.2246	0.4565
1.3025	16.0	56	1.3046	0.4565
1.3025	16.86	59	1.2177	0.4565
1.2981	18.0	63	1.2339	0.4565
1.2981	18.86	66	1.3139	0.4565
1.2765	20.0	70	1.2116	0.4565
1.2765	20.86	73	1.2284	0.3261
1.2765	22.0	77	1.2246	0.4565
1.2074	22.86	80	1.2631	0.4565
1.2074	24.0	84	1.2092	0.4565
1.2074	24.86	87	1.2147	0.4565
1.2048	26.0	91	1.2121	0.4565
1.2048	26.86	94	1.2156	0.4565
1.2048	28.0	98	1.2249	0.4565
1.2068	28.86	101	1.2159	0.4565
1.2068	30.0	105	1.2108	0.4565
1.2068	30.86	108	1.2116	0.4565
1.1961	32.0	112	1.2078	0.4565
1.1961	32.86	115	1.2070	0.4565
1.1961	34.0	119	1.2072	0.4565
1.1999	34.29	120	1.2072	0.4565

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu118
Datasets 2.16.1
Tokenizers 0.15.0