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: 0.3295
• 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	0.95	5	0.6389	0.0133
No log	1.9	10	0.5645	0.0
No log	2.86	15	0.4792	0.0
No log	4.0	21	0.4224	0.0
No log	4.95	26	0.3652	0.0
No log	5.9	31	0.3357	0.0
No log	6.86	36	0.2953	0.0
No log	8.0	42	0.2909	0.0133
No log	8.95	47	0.2937	0.7333
No log	9.9	52	0.2718	0.68
No log	10.86	57	0.2673	0.64
No log	12.0	63	0.3019	0.8667
No log	12.95	68	0.2945	0.4667
No log	13.9	73	0.2669	0.6667
No log	14.86	78	0.2504	0.7467
No log	16.0	84	0.2380	0.64
No log	16.95	89	0.2525	0.64
No log	17.9	94	0.2648	0.7467
No log	18.86	99	0.2711	0.7067
No log	20.0	105	0.2747	0.64
No log	20.95	110	0.2772	0.64
No log	21.9	115	0.3000	0.6267
No log	22.86	120	0.2871	0.5733
No log	24.0	126	0.3025	0.6667
No log	24.95	131	0.3317	0.5867
No log	25.9	136	0.3171	0.5467
No log	26.86	141	0.3322	0.64
No log	28.0	147	0.3207	0.6533
No log	28.95	152	0.3492	0.5733
No log	29.9	157	0.2965	0.68
No log	30.86	162	0.3256	0.72
No log	32.0	168	0.3460	0.6267
No log	32.95	173	0.3118	0.7067
No log	33.9	178	0.3656	0.6933
No log	34.86	183	0.3111	0.5867
No log	36.0	189	0.3119	0.6667
No log	36.95	194	0.3524	0.72
No log	37.9	199	0.3457	0.5333
No log	38.86	204	0.3460	0.56
No log	40.0	210	0.3518	0.6933
No log	40.95	215	0.2948	0.5867
No log	41.9	220	0.3640	0.5867
No log	42.86	225	0.3408	0.6133
No log	44.0	231	0.3350	0.6
No log	44.95	236	0.3832	0.72
No log	45.9	241	0.3298	0.6667
No log	46.86	246	0.3557	0.64
No log	48.0	252	0.3832	0.64
No log	48.95	257	0.3317	0.6533
No log	49.9	262	0.3980	0.6933
No log	50.86	267	0.3295	0.6533

Framework versions

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