metadata

license: apache-2.0
base_model: microsoft/beit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - precision
  - recall
model-index:
  - name: beit-base-patch16-224
    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: 0.85
          - name: Precision
            type: precision
            value: 0.8455590062111802
          - name: Recall
            type: recall
            value: 0.85

beit-base-patch16-224

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: 0.4871
Accuracy: 0.85
Precision: 0.8456
Recall: 0.85
F1 Score: 0.8464

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: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 Score
No log	1.0	4	0.5784	0.7333	0.5378	0.7333	0.6205
No log	2.0	8	0.5813	0.7375	0.7030	0.7375	0.6441
No log	3.0	12	0.5486	0.7417	0.7297	0.7417	0.7343
No log	4.0	16	0.5394	0.7542	0.7333	0.7542	0.7370
No log	5.0	20	0.5067	0.775	0.7658	0.775	0.7321
No log	6.0	24	0.5542	0.7958	0.7966	0.7958	0.7613
No log	7.0	28	0.4753	0.7958	0.7834	0.7958	0.7758
0.5325	8.0	32	0.5265	0.7792	0.7661	0.7792	0.7448
0.5325	9.0	36	0.4789	0.8208	0.8134	0.8208	0.8067
0.5325	10.0	40	0.4939	0.7875	0.7932	0.7875	0.7900
0.5325	11.0	44	0.4917	0.8042	0.8032	0.8042	0.8037
0.5325	12.0	48	0.5001	0.8083	0.8019	0.8083	0.8041
0.5325	13.0	52	0.4742	0.8	0.7897	0.8	0.7915
0.5325	14.0	56	0.5439	0.7875	0.8037	0.7875	0.7932
0.3381	15.0	60	0.5436	0.8333	0.8265	0.8333	0.8263
0.3381	16.0	64	0.4989	0.8375	0.8312	0.8375	0.8288
0.3381	17.0	68	0.4949	0.8333	0.8282	0.8333	0.8296
0.3381	18.0	72	0.4709	0.8292	0.8283	0.8292	0.8287
0.3381	19.0	76	0.4680	0.8167	0.8133	0.8167	0.8147
0.3381	20.0	80	0.5053	0.8417	0.8362	0.8417	0.8371
0.3381	21.0	84	0.5480	0.8458	0.8459	0.8458	0.8322
0.3381	22.0	88	0.4548	0.8542	0.8512	0.8542	0.8522
0.2076	23.0	92	0.4891	0.8458	0.8407	0.8458	0.8376
0.2076	24.0	96	0.4981	0.85	0.8486	0.85	0.8492
0.2076	25.0	100	0.4993	0.8458	0.8426	0.8458	0.8438
0.2076	26.0	104	0.5026	0.8542	0.8503	0.8542	0.8514
0.2076	27.0	108	0.4944	0.8542	0.8522	0.8542	0.8530
0.2076	28.0	112	0.4821	0.8542	0.8549	0.8542	0.8545
0.2076	29.0	116	0.4714	0.8583	0.8559	0.8583	0.8568
0.138	30.0	120	0.4705	0.8583	0.8559	0.8583	0.8568

Framework versions

Transformers 4.33.3
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3