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_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: 0.8048780487804879

hushem_1x_beit_base_adamax_00001_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: 0.5012
• Accuracy: 0.8049

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	1.3123	0.3415
1.3417	2.0	12	1.1955	0.5610
1.3417	3.0	18	1.0858	0.6098
0.9733	4.0	24	0.9881	0.6098
0.7755	5.0	30	0.9019	0.6829
0.7755	6.0	36	0.8266	0.7073
0.5866	7.0	42	0.7789	0.7073
0.5866	8.0	48	0.7323	0.7561
0.4247	9.0	54	0.6965	0.7805
0.3366	10.0	60	0.6709	0.8049
0.3366	11.0	66	0.6414	0.8049
0.2406	12.0	72	0.6309	0.8049
0.2406	13.0	78	0.6283	0.8049
0.1807	14.0	84	0.5831	0.7805
0.1504	15.0	90	0.5587	0.8049
0.1504	16.0	96	0.5622	0.8049
0.1175	17.0	102	0.5998	0.8049
0.1175	18.0	108	0.5569	0.8049
0.1107	19.0	114	0.5178	0.7805
0.0834	20.0	120	0.5243	0.7805
0.0834	21.0	126	0.5491	0.8293
0.0685	22.0	132	0.5414	0.8049
0.0685	23.0	138	0.5016	0.8049
0.0622	24.0	144	0.5127	0.8049
0.0553	25.0	150	0.5123	0.8049
0.0553	26.0	156	0.5115	0.8293
0.0606	27.0	162	0.5050	0.8293
0.0606	28.0	168	0.4920	0.8293
0.0617	29.0	174	0.4931	0.8049
0.042	30.0	180	0.5002	0.8049
0.042	31.0	186	0.5113	0.8049
0.0437	32.0	192	0.5113	0.8049
0.0437	33.0	198	0.5177	0.8049
0.0387	34.0	204	0.5166	0.8049
0.0462	35.0	210	0.5121	0.8049
0.0462	36.0	216	0.4996	0.8049
0.0301	37.0	222	0.4928	0.8049
0.0301	38.0	228	0.4955	0.8049
0.0363	39.0	234	0.4997	0.8049
0.0291	40.0	240	0.5021	0.8049
0.0291	41.0	246	0.5012	0.8049
0.0319	42.0	252	0.5012	0.8049
0.0319	43.0	258	0.5012	0.8049
0.0305	44.0	264	0.5012	0.8049
0.0298	45.0	270	0.5012	0.8049
0.0298	46.0	276	0.5012	0.8049
0.0437	47.0	282	0.5012	0.8049
0.0437	48.0	288	0.5012	0.8049
0.03	49.0	294	0.5012	0.8049
0.0346	50.0	300	0.5012	0.8049

Framework versions

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