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-RHS-DA
    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: 0.616822429906542

SW2-RHS-DA

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: 2.1002
• Accuracy: 0.6168

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: 4e-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
1.302	0.98	22	2.8474	0.4112
0.6224	2.0	45	1.4650	0.4112
0.3905	2.98	67	1.4865	0.4112
0.1416	4.0	90	0.9453	0.5981
0.1116	4.98	112	0.9801	0.5514
0.0866	6.0	135	1.5201	0.6075
0.0579	6.98	157	1.7234	0.5981
0.0667	8.0	180	1.9649	0.5981
0.0664	8.98	202	1.9505	0.5981
0.0742	10.0	225	1.9448	0.5981
0.0558	10.98	247	1.9545	0.5981
0.0475	12.0	270	2.1516	0.5888
0.114	12.98	292	2.1002	0.6168
0.051	14.0	315	2.2643	0.5981
0.0318	14.98	337	2.3468	0.5981
0.0673	16.0	360	2.3341	0.6075
0.0566	16.98	382	2.3191	0.6075
0.0543	18.0	405	2.2807	0.6168
0.0458	18.98	427	2.2148	0.6168
0.0421	20.0	450	2.5468	0.5888
0.0139	20.98	472	2.2408	0.6168
0.012	22.0	495	2.5689	0.5888
0.017	22.98	517	2.5485	0.6168
0.0529	24.0	540	2.6746	0.6075
0.0414	24.98	562	2.7693	0.5888
0.0158	26.0	585	2.7447	0.5888
0.0205	26.98	607	2.8566	0.5888
0.0205	28.0	630	2.8469	0.5888
0.006	28.98	652	2.9508	0.5888
0.0061	30.0	675	3.0560	0.5888
0.0227	30.98	697	3.0431	0.5888
0.034	32.0	720	3.0497	0.5888
0.0039	32.98	742	3.0936	0.5888
0.0031	34.0	765	3.1158	0.5888
0.0118	34.98	787	3.1093	0.5888
0.0045	36.0	810	3.0885	0.5888
0.0168	36.98	832	3.0814	0.5888
0.0071	38.0	855	3.0733	0.5888
0.0238	38.98	877	3.0825	0.5888
0.0072	39.11	880	3.0828	0.5888

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu118
• Datasets 2.16.1
• Tokenizers 0.15.0