README.md

metadata

license: apache-2.0
base_model: microsoft/swinv2-tiny-patch4-window8-256
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: SW2-DMAE
    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

SW2-DMAE

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: 1.6739
• 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: 5e-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
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.86	3	7.9394	0.1087
No log	2.0	7	7.8979	0.1087
7.935	2.86	10	7.7672	0.1087
7.935	4.0	14	7.2197	0.1087
7.935	4.86	17	6.5661	0.1087
7.0143	6.0	21	5.7304	0.1087
7.0143	6.86	24	5.1184	0.1087
7.0143	8.0	28	4.3526	0.1087
4.9972	8.86	31	3.8117	0.1087
4.9972	10.0	35	3.1518	0.1087
4.9972	10.86	38	2.7125	0.1087
3.3803	12.0	42	2.2254	0.1087
3.3803	12.86	45	1.9450	0.1087
3.3803	14.0	49	1.6739	0.4565
2.0759	14.86	52	1.5299	0.4565
2.0759	16.0	56	1.3876	0.4565
2.0759	16.86	59	1.3059	0.4565
1.4466	18.0	63	1.2341	0.4565
1.4466	18.86	66	1.2120	0.4565
1.2349	20.0	70	1.2096	0.4565
1.2349	20.86	73	1.2118	0.4565
1.2349	22.0	77	1.2114	0.4565
1.1854	22.86	80	1.2141	0.4565
1.1854	24.0	84	1.2117	0.4565
1.1854	24.86	87	1.2102	0.4565
1.1878	26.0	91	1.2076	0.4565
1.1878	26.86	94	1.2083	0.4565
1.1878	28.0	98	1.2130	0.4565
1.1986	28.86	101	1.2069	0.4565
1.1986	30.0	105	1.2058	0.4565
1.1986	30.86	108	1.2070	0.4565
1.182	32.0	112	1.2075	0.4565
1.182	32.86	115	1.2074	0.4565
1.182	34.0	119	1.2072	0.4565
1.2064	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