README.md

metadata

library_name: transformers
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-finetuned-lf-invalidation
    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: 0.676595744680851

swinv2-tiny-patch4-window8-256-finetuned-lf-invalidation

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: 0.7236
• Accuracy: 0.6766

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5608	0.9796	12	0.6779	0.5532
0.5249	1.9592	24	0.5234	0.7617
0.4404	2.9388	36	0.5121	0.7766
0.3893	4.0	49	0.3981	0.8128
0.4083	4.9796	61	0.5870	0.6447
0.3725	5.9592	73	0.4991	0.7553
0.3909	6.9388	85	0.4062	0.8426
0.3799	8.0	98	0.5115	0.7574
0.3332	8.9796	110	0.4470	0.8277
0.3108	9.9592	122	0.3451	0.8681
0.308	10.9388	134	0.5822	0.7511
0.3699	12.0	147	0.4653	0.8106
0.2945	12.9796	159	0.4171	0.8426
0.2934	13.9592	171	0.4366	0.8234
0.2719	14.9388	183	0.5905	0.7638
0.3287	16.0	196	0.6654	0.7234
0.271	16.9796	208	0.6328	0.7447
0.3018	17.9592	220	0.4671	0.8255
0.2763	18.9388	232	0.6032	0.7468
0.2834	20.0	245	0.7016	0.7
0.2653	20.9796	257	0.4089	0.8468
0.2666	21.9592	269	0.7905	0.6447
0.2941	22.9388	281	0.6064	0.7553
0.2792	24.0	294	0.7444	0.7085
0.2019	24.9796	306	0.7595	0.7170
0.2552	25.9592	318	1.0296	0.5660
0.2451	26.9388	330	0.5999	0.7340
0.2126	28.0	343	0.5730	0.7660
0.2214	28.9796	355	0.9756	0.5809
0.2633	29.9592	367	0.4134	0.8404
0.2427	30.9388	379	0.8228	0.6362
0.2405	32.0	392	0.5279	0.7723
0.2078	32.9796	404	0.6581	0.6979
0.2201	33.9592	416	0.9132	0.5745
0.2481	34.9388	428	0.9526	0.5617
0.248	36.0	441	0.8979	0.5553
0.2209	36.9796	453	0.8351	0.5915
0.2253	37.9592	465	0.6744	0.6851
0.2447	38.9388	477	0.7794	0.6404
0.2049	40.0	490	0.6136	0.7468
0.1965	40.9796	502	0.6582	0.7340
0.2638	41.9592	514	0.7487	0.6766
0.2297	42.9388	526	0.7282	0.6702
0.2163	44.0	539	0.5713	0.7511
0.2016	44.9796	551	0.5994	0.7319
0.1739	45.9592	563	0.6865	0.6915
0.2497	46.9388	575	0.6901	0.6957
0.2293	48.0	588	0.7150	0.6851
0.2237	48.9796	600	0.7236	0.6766

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1