README.md

metadata

license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - image-classification
  - generated_from_trainer
datasets:
  - renovation
metrics:
  - accuracy
model-index:
  - name: renovation
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: beans
          type: renovation
          config: default
          split: validation
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7219562243502052

renovation

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the beans dataset. It achieves the following results on the evaluation set:

• Loss: 0.6830
• Accuracy: 0.7220

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: 0.0002
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0475	0.07	100	1.0332	0.5824
0.8651	0.14	200	0.9322	0.6204
1.0022	0.21	300	1.2150	0.5147
1.0636	0.27	400	0.9523	0.6252
0.8311	0.34	500	0.8440	0.6556
0.88	0.41	600	0.8707	0.6495
0.8881	0.48	700	0.8903	0.6334
0.7522	0.55	800	0.8479	0.6577
0.798	0.62	900	0.7739	0.6843
0.7317	0.68	1000	0.7856	0.6795
0.8372	0.75	1100	0.8884	0.6354
0.6629	0.82	1200	0.7573	0.6871
0.7767	0.89	1300	0.7543	0.6860
0.9246	0.96	1400	0.7896	0.6635
0.5026	1.03	1500	0.7872	0.6813
0.7599	1.1	1600	0.7861	0.6758
0.5764	1.16	1700	0.8088	0.6802
0.4329	1.23	1800	0.7281	0.7059
0.6271	1.3	1900	0.7291	0.7117
0.5498	1.37	2000	0.7745	0.7059
0.5247	1.44	2100	0.8002	0.6891
0.4891	1.51	2200	0.7014	0.7100
0.5211	1.57	2300	0.7725	0.6864
0.659	1.64	2400	0.7477	0.7086
0.4878	1.71	2500	0.7129	0.7052
0.4941	1.78	2600	0.6830	0.7220
0.4648	1.85	2700	0.7182	0.7028
0.5501	1.92	2800	0.7191	0.7144
0.5491	1.98	2900	0.7132	0.7155
0.2373	2.05	3000	0.7831	0.7096
0.2756	2.12	3100	0.7965	0.7247
0.2299	2.19	3200	0.8241	0.7220
0.2323	2.26	3300	0.8286	0.7110
0.1979	2.33	3400	0.7993	0.7302
0.2507	2.4	3500	0.8477	0.7189
0.205	2.46	3600	0.8197	0.7124
0.35	2.53	3700	0.8348	0.7127
0.3372	2.6	3800	0.8999	0.7199
0.1968	2.67	3900	0.8263	0.7274
0.1443	2.74	4000	0.8704	0.7244
0.1933	2.81	4100	0.8270	0.7244
0.2044	2.87	4200	0.8323	0.7274
0.2709	2.94	4300	0.8494	0.7295
0.1021	3.01	4400	0.8573	0.7336
0.0393	3.08	4500	0.9333	0.7377
0.0973	3.15	4600	0.9646	0.7336
0.0317	3.22	4700	0.9820	0.7336
0.0458	3.29	4800	1.0716	0.7326
0.164	3.35	4900	1.0889	0.7312
0.0578	3.42	5000	1.1011	0.7312
0.0563	3.49	5100	1.1010	0.7356
0.0318	3.56	5200	1.0923	0.7343
0.0255	3.63	5300	1.1156	0.7332
0.0169	3.7	5400	1.1050	0.7415
0.0629	3.76	5500	1.1132	0.7373
0.0627	3.83	5600	1.1110	0.7380
0.0078	3.9	5700	1.1117	0.7350
0.027	3.97	5800	1.1201	0.7343

Framework versions

• Transformers 4.39.1
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2