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-5e-2
    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-5e-2

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: 8.3393
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: 0.05
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
num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.86	3	22.4107	0.3261
No log	2.0	7	11.2632	0.3261
15.4384	2.86	10	8.3393	0.4565
15.4384	4.0	14	2.2808	0.1087
15.4384	4.86	17	2.9784	0.3261
4.9288	6.0	21	2.4757	0.3261
4.9288	6.86	24	2.5934	0.1087
4.9288	8.0	28	2.3187	0.4565
2.426	8.86	31	1.7270	0.3261
2.426	10.0	35	1.4868	0.4565
2.426	10.86	38	1.3056	0.3261
1.5653	12.0	42	1.2237	0.4565
1.5653	12.86	45	1.4644	0.4565
1.5653	14.0	49	1.2427	0.4565
1.4578	14.86	52	1.2474	0.3261
1.4578	16.0	56	1.3441	0.4565
1.4578	16.86	59	1.3480	0.3261
1.3492	18.0	63	1.2835	0.4565
1.3492	18.86	66	1.3153	0.4565
1.3089	20.0	70	1.2387	0.4565
1.3089	20.86	73	1.2538	0.3261
1.3089	22.0	77	1.2386	0.4565
1.2345	22.86	80	1.2612	0.3261
1.2345	24.0	84	1.2125	0.4565
1.2345	24.86	87	1.2079	0.4565
1.2105	26.0	91	1.2139	0.4565
1.2105	26.86	94	1.2171	0.4565
1.2105	28.0	98	1.2183	0.4565
1.2095	28.86	101	1.2094	0.4565
1.2095	30.0	105	1.2061	0.4565
1.2095	30.86	108	1.2069	0.4565
1.203	32.0	112	1.2073	0.4565
1.203	32.86	115	1.2072	0.4565
1.203	34.0	119	1.2067	0.4565
1.1996	34.29	120	1.2066	0.4565

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu118
Datasets 2.16.1
Tokenizers 0.15.0