metadata

license: apache-2.0
base_model: microsoft/beit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: beit-base-patch16-224-OT-alt
    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: 0.8387096774193549

beit-base-patch16-224-OT-alt

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

Loss: 0.5389
Accuracy: 0.8387

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: 3.8e-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	0.91	5	1.7603	0.1452
1.7693	2.0	11	1.6916	0.1452
1.7693	2.91	16	1.5752	0.1452
1.6261	4.0	22	1.4015	0.1452
1.6261	4.91	27	1.2890	0.1452
1.3534	6.0	33	1.2128	0.3710
1.3534	6.91	38	1.1418	0.4032
1.1661	8.0	44	1.0727	0.4677
1.1661	8.91	49	1.0909	0.4032
1.0344	10.0	55	0.9719	0.6129
0.9604	10.91	60	0.9923	0.6452
0.9604	12.0	66	0.9554	0.6290
0.8477	12.91	71	0.9156	0.6774
0.8477	14.0	77	0.8339	0.7097
0.7727	14.91	82	0.7851	0.7258
0.7727	16.0	88	0.7994	0.7258
0.6714	16.91	93	0.8246	0.6290
0.6714	18.0	99	0.7389	0.7097
0.6143	18.91	104	0.8202	0.6452
0.5398	20.0	110	0.6295	0.7742
0.5398	20.91	115	0.6736	0.7581
0.4958	22.0	121	0.6218	0.7903
0.4958	22.91	126	0.6401	0.7742
0.4561	24.0	132	0.6640	0.7258
0.4561	24.91	137	0.6009	0.7742
0.4149	26.0	143	0.5619	0.8065
0.4149	26.91	148	0.5118	0.8065
0.3965	28.0	154	0.5682	0.8065
0.3965	28.91	159	0.5515	0.8065
0.4219	30.0	165	0.7045	0.7097
0.3939	30.91	170	0.5712	0.8065
0.3939	32.0	176	0.5857	0.8065
0.3598	32.91	181	0.5539	0.8065
0.3598	34.0	187	0.5471	0.8226
0.3613	34.91	192	0.5408	0.8226
0.3613	36.0	198	0.5389	0.8387
0.3748	36.36	200	0.5390	0.8387

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu118
Datasets 2.16.1
Tokenizers 0.15.0