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: hushem_5x_beit_base_adamax_0001_fold5
    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: 0.8292682926829268

hushem_5x_beit_base_adamax_0001_fold5

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: 0.9920
• Accuracy: 0.8293

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
0.641	1.0	28	0.5863	0.7317
0.1363	2.0	56	0.5085	0.8049
0.0403	3.0	84	0.8067	0.8049
0.0158	4.0	112	0.6196	0.7805
0.0024	5.0	140	0.5093	0.9024
0.0017	6.0	168	0.8498	0.8293
0.0016	7.0	196	0.8351	0.8537
0.0006	8.0	224	0.8986	0.8293
0.0004	9.0	252	0.5488	0.8780
0.0018	10.0	280	0.6108	0.8780
0.0004	11.0	308	0.7112	0.8537
0.0001	12.0	336	0.7806	0.8537
0.0001	13.0	364	0.8893	0.8780
0.0001	14.0	392	0.9596	0.8293
0.005	15.0	420	0.6813	0.9024
0.0001	16.0	448	0.9535	0.7561
0.0002	17.0	476	0.6993	0.8537
0.0001	18.0	504	0.6415	0.8537
0.0001	19.0	532	0.6851	0.8293
0.0	20.0	560	0.6721	0.8537
0.0001	21.0	588	0.6658	0.8537
0.0001	22.0	616	0.6422	0.8537
0.0	23.0	644	0.6457	0.8537
0.0001	24.0	672	0.6659	0.8537
0.0001	25.0	700	0.6825	0.8537
0.0	26.0	728	0.7334	0.8293
0.0	27.0	756	0.7447	0.8293
0.0	28.0	784	0.7524	0.8293
0.0001	29.0	812	0.7681	0.8293
0.0	30.0	840	0.7829	0.8293
0.0001	31.0	868	0.9622	0.8293
0.0002	32.0	896	1.1091	0.8293
0.0	33.0	924	1.0337	0.8293
0.0	34.0	952	0.9846	0.8293
0.0	35.0	980	0.9615	0.8293
0.0001	36.0	1008	0.9612	0.8293
0.0001	37.0	1036	0.9494	0.8293
0.0001	38.0	1064	0.9364	0.8293
0.0	39.0	1092	0.9014	0.8293
0.0002	40.0	1120	0.8887	0.8293
0.0001	41.0	1148	0.9321	0.8293
0.0	42.0	1176	0.9557	0.8293
0.0	43.0	1204	1.0018	0.8293
0.0	44.0	1232	1.0026	0.8293
0.0006	45.0	1260	0.9880	0.8293
0.0001	46.0	1288	0.9929	0.8293
0.0001	47.0	1316	0.9924	0.8293
0.0	48.0	1344	0.9920	0.8293
0.0	49.0	1372	0.9920	0.8293
0.0	50.0	1400	0.9920	0.8293

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0