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_1x_beit_base_adamax_00001_fold4
    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: 0.8571428571428571

hushem_1x_beit_base_adamax_00001_fold4

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: 0.3525
• Accuracy: 0.8571

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: 1e-05
• 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
No log	1.0	6	1.2487	0.4524
1.3325	2.0	12	1.1186	0.5476
1.3325	3.0	18	1.0007	0.6429
0.9723	4.0	24	0.9193	0.6429
0.7474	5.0	30	0.8604	0.6429
0.7474	6.0	36	0.7650	0.7143
0.5775	7.0	42	0.6832	0.7619
0.5775	8.0	48	0.6554	0.7381
0.4415	9.0	54	0.6117	0.7381
0.3035	10.0	60	0.5783	0.7857
0.3035	11.0	66	0.5572	0.7857
0.2259	12.0	72	0.5149	0.8095
0.2259	13.0	78	0.4887	0.8571
0.1918	14.0	84	0.4779	0.8095
0.1546	15.0	90	0.4660	0.8095
0.1546	16.0	96	0.4640	0.8333
0.1177	17.0	102	0.4361	0.9048
0.1177	18.0	108	0.4152	0.8810
0.1109	19.0	114	0.4213	0.9048
0.0885	20.0	120	0.4055	0.9048
0.0885	21.0	126	0.3889	0.8810
0.0665	22.0	132	0.3826	0.8810
0.0665	23.0	138	0.3817	0.9048
0.0682	24.0	144	0.3714	0.9048
0.055	25.0	150	0.3759	0.8810
0.055	26.0	156	0.3597	0.9048
0.0545	27.0	162	0.3543	0.9048
0.0545	28.0	168	0.3635	0.8810
0.0561	29.0	174	0.3559	0.8810
0.0403	30.0	180	0.3405	0.9048
0.0403	31.0	186	0.3424	0.9048
0.048	32.0	192	0.3445	0.8810
0.048	33.0	198	0.3419	0.8810
0.0375	34.0	204	0.3424	0.8810
0.0438	35.0	210	0.3440	0.8810
0.0438	36.0	216	0.3465	0.8571
0.0357	37.0	222	0.3469	0.8571
0.0357	38.0	228	0.3472	0.8571
0.0346	39.0	234	0.3498	0.8571
0.0356	40.0	240	0.3512	0.8571
0.0356	41.0	246	0.3523	0.8571
0.042	42.0	252	0.3525	0.8571
0.042	43.0	258	0.3525	0.8571
0.033	44.0	264	0.3525	0.8571
0.0336	45.0	270	0.3525	0.8571
0.0336	46.0	276	0.3525	0.8571
0.0414	47.0	282	0.3525	0.8571
0.0414	48.0	288	0.3525	0.8571
0.0414	49.0	294	0.3525	0.8571
0.0317	50.0	300	0.3525	0.8571

Framework versions

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