README.md

metadata

license: apache-2.0
base_model: microsoft/beit-base-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_5x_beit_base_sgd_0001_fold2
    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: 0.8036605657237936

smids_5x_beit_base_sgd_0001_fold2

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: 0.4724
• Accuracy: 0.8037

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• 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
1.0748	1.0	375	1.2296	0.3677
1.052	2.0	750	1.1460	0.3943
0.9644	3.0	1125	1.0650	0.4326
0.8415	4.0	1500	0.9877	0.4942
0.786	5.0	1875	0.9192	0.5557
0.7841	6.0	2250	0.8580	0.6323
0.7782	7.0	2625	0.8061	0.6689
0.7211	8.0	3000	0.7620	0.6889
0.6883	9.0	3375	0.7245	0.7138
0.6641	10.0	3750	0.6925	0.7371
0.6683	11.0	4125	0.6665	0.7404
0.7093	12.0	4500	0.6445	0.7454
0.5818	13.0	4875	0.6259	0.7604
0.5841	14.0	5250	0.6091	0.7604
0.5811	15.0	5625	0.5946	0.7637
0.5799	16.0	6000	0.5819	0.7704
0.5841	17.0	6375	0.5719	0.7720
0.5531	18.0	6750	0.5621	0.7770
0.5613	19.0	7125	0.5532	0.7820
0.5733	20.0	7500	0.5445	0.7837
0.538	21.0	7875	0.5380	0.7887
0.5305	22.0	8250	0.5326	0.7903
0.5558	23.0	8625	0.5259	0.7887
0.5149	24.0	9000	0.5202	0.7903
0.5317	25.0	9375	0.5157	0.7903
0.5391	26.0	9750	0.5118	0.7937
0.557	27.0	10125	0.5073	0.7920
0.4911	28.0	10500	0.5033	0.7970
0.4985	29.0	10875	0.5001	0.7937
0.5262	30.0	11250	0.4968	0.7937
0.4712	31.0	11625	0.4944	0.7970
0.5163	32.0	12000	0.4912	0.7953
0.4489	33.0	12375	0.4890	0.7987
0.4565	34.0	12750	0.4870	0.8003
0.484	35.0	13125	0.4849	0.7987
0.4879	36.0	13500	0.4832	0.8020
0.48	37.0	13875	0.4815	0.8053
0.4581	38.0	14250	0.4797	0.8053
0.4627	39.0	14625	0.4783	0.8070
0.475	40.0	15000	0.4772	0.8053
0.4851	41.0	15375	0.4762	0.8037
0.4434	42.0	15750	0.4753	0.8037
0.4381	43.0	16125	0.4746	0.8037
0.506	44.0	16500	0.4740	0.8037
0.4383	45.0	16875	0.4734	0.8037
0.4915	46.0	17250	0.4730	0.8037
0.4925	47.0	17625	0.4727	0.8037
0.4423	48.0	18000	0.4725	0.8037
0.4853	49.0	18375	0.4724	0.8037
0.5129	50.0	18750	0.4724	0.8037

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2