README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - image_folder
metrics:
  - accuracy
model-index:
  - name: Beit-for-rice-disease
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: image_folder
          type: image_folder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.95

Beit-for-rice-disease

This model is a fine-tuned version of microsoft/beit-base-patch16-224-pt22k-ft22k on the image_folder dataset. It achieves the following results on the evaluation set:

• Loss: 0.1687
• Accuracy: 0.95

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: 5e-05
• 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.4305	0.16	10	2.0142	0.5
1.804	0.32	20	1.5097	0.5446
1.2088	0.48	30	1.0185	0.6786
0.9168	0.63	40	0.8841	0.7768
0.7893	0.79	50	0.7512	0.7768
0.6253	0.95	60	0.9253	0.7411
0.4949	1.11	70	0.5854	0.8214
0.3272	1.27	80	0.5379	0.8393
0.3753	1.43	90	0.5194	0.8571
0.3378	1.59	100	0.3831	0.8661
0.3643	1.75	110	0.4066	0.8661
0.3413	1.9	120	0.4268	0.875
0.3278	2.06	130	0.4754	0.8482
0.1784	2.22	140	0.4979	0.8839
0.2002	2.38	150	0.3606	0.8929
0.1224	2.54	160	0.3852	0.8839
0.2458	2.7	170	0.3492	0.9107
0.1816	2.86	180	0.3506	0.8839
0.2194	3.02	190	0.3255	0.9018
0.1582	3.17	200	0.4486	0.875
0.2072	3.33	210	0.3760	0.8929
0.1317	3.49	220	0.3021	0.9286
0.1223	3.65	230	0.3867	0.9018
0.1543	3.81	240	0.2640	0.9286
0.1256	3.97	250	0.2925	0.9196
0.095	4.13	260	0.2909	0.9107
0.0578	4.29	270	0.2825	0.9196
0.0428	4.44	280	0.2826	0.9196
0.1743	4.6	290	0.2746	0.9196
0.0659	4.76	300	0.3457	0.8929
0.0871	4.92	310	0.2902	0.9196
0.0726	5.08	320	0.2810	0.9196
0.0556	5.24	330	0.3244	0.9107
0.0629	5.4	340	0.2812	0.9286
0.0517	5.56	350	0.2638	0.9464
0.078	5.71	360	0.2580	0.9554
0.0376	5.87	370	0.2375	0.9464
0.0262	6.03	380	0.2274	0.9464
0.0582	6.19	390	0.2254	0.9643
0.0215	6.35	400	0.2545	0.9375
0.0379	6.51	410	0.2925	0.9286
0.0664	6.67	420	0.2842	0.9196
0.0375	6.83	430	0.2643	0.9375
0.0088	6.98	440	0.2401	0.9375
0.0299	7.14	450	0.2634	0.9286
0.0199	7.3	460	0.2800	0.9286
0.0182	7.46	470	0.2996	0.9107
0.0199	7.62	480	0.2813	0.9286
0.0189	7.78	490	0.2571	0.9286
0.053	7.94	500	0.2260	0.9196
0.0448	8.1	510	0.2112	0.9464
0.0072	8.25	520	0.2133	0.9286
0.0112	8.41	530	0.2149	0.9464
0.021	8.57	540	0.2189	0.9286
0.0073	8.73	550	0.2397	0.9375
0.0128	8.89	560	0.2578	0.9375
0.005	9.05	570	0.2578	0.9375
0.0246	9.21	580	0.2485	0.9286
0.0073	9.37	590	0.2460	0.9286
0.005	9.52	600	0.2429	0.9286
0.0083	9.68	610	0.2362	0.9286
0.0114	9.84	620	0.2338	0.9286
0.0031	10.0	630	0.2327	0.9286

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3