README.md

metadata

license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - recall
model-index:
  - name: vca
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: test
          args: default
        metrics:
          - name: Recall
            type: recall
            value: 0.6533333333333333

vca

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.2071
• Recall: 0.6533

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: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Recall
No log	1.0	9	0.6362	0.3333
No log	2.0	18	0.4641	0.0
No log	3.0	27	0.3251	0.0
No log	4.0	36	0.2605	0.0
No log	5.0	45	0.2100	0.0
No log	6.0	54	0.1943	0.08
No log	7.0	63	0.1986	0.64
No log	8.0	72	0.1856	0.6933
No log	9.0	81	0.1654	0.6933
No log	10.0	90	0.1593	0.72
No log	11.0	99	0.1638	0.68
No log	12.0	108	0.1732	0.6933
No log	13.0	117	0.1748	0.56
No log	14.0	126	0.1792	0.6533
No log	15.0	135	0.1743	0.84
No log	16.0	144	0.1760	0.5733
No log	17.0	153	0.1641	0.6
No log	18.0	162	0.1558	0.76
No log	19.0	171	0.2121	0.7867
No log	20.0	180	0.1765	0.56
No log	21.0	189	0.1802	0.7733
No log	22.0	198	0.1729	0.7467
No log	23.0	207	0.2004	0.48
No log	24.0	216	0.1794	0.72
No log	25.0	225	0.2185	0.7867
No log	26.0	234	0.2115	0.8533
No log	27.0	243	0.1999	0.7067
No log	28.0	252	0.1900	0.5467
No log	29.0	261	0.2158	0.72
No log	30.0	270	0.2515	0.8533
No log	31.0	279	0.2322	0.7733
No log	32.0	288	0.2024	0.8
No log	33.0	297	0.2342	0.76
No log	34.0	306	0.2205	0.7467
No log	35.0	315	0.1820	0.7067
No log	36.0	324	0.2169	0.68
No log	37.0	333	0.2170	0.6133
No log	38.0	342	0.1767	0.68
No log	39.0	351	0.2326	0.8133
No log	40.0	360	0.2386	0.76
No log	41.0	369	0.2431	0.68
No log	42.0	378	0.2160	0.6933
No log	43.0	387	0.2234	0.76
No log	44.0	396	0.2491	0.7467
No log	45.0	405	0.2342	0.6933
No log	46.0	414	0.2124	0.7333
No log	47.0	423	0.2602	0.6533
No log	48.0	432	0.2702	0.6133
No log	49.0	441	0.2258	0.6533
No log	50.0	450	0.2158	0.64
No log	51.0	459	0.2071	0.6533

Framework versions

• Transformers 4.34.1
• Pytorch 2.0.1+cu117
• Datasets 2.14.6
• Tokenizers 0.14.1