README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: vit-base-patch16-224-Soybean_11-46
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9305555555555556

vit-base-patch16-224-Soybean_11-46

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

• Loss: 0.2058
• Accuracy: 0.9306

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: 60
• eval_batch_size: 60
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 240
• 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
1.3661	1.0	11	1.3698	0.5069
0.9979	2.0	22	0.9817	0.6632
0.6746	3.0	33	0.7423	0.7396
0.6364	4.0	44	0.6075	0.7569
0.5425	5.0	55	0.5500	0.7951
0.5001	6.0	66	0.4883	0.8160
0.3519	7.0	77	0.4539	0.8264
0.4421	8.0	88	0.4483	0.8194
0.3207	9.0	99	0.3785	0.8438
0.3682	10.0	110	0.3385	0.8646
0.2642	11.0	121	0.3827	0.8403
0.3444	12.0	132	0.3462	0.8507
0.2423	13.0	143	0.3170	0.8681
0.3168	14.0	154	0.3168	0.8715
0.2781	15.0	165	0.3323	0.8333
0.2411	16.0	176	0.3200	0.8715
0.2276	17.0	187	0.3296	0.875
0.192	18.0	198	0.3119	0.8854
0.1612	19.0	209	0.3647	0.875
0.1084	20.0	220	0.2641	0.8993
0.2099	21.0	231	0.2807	0.8958
0.1666	22.0	242	0.2595	0.9097
0.1355	23.0	253	0.2735	0.8924
0.1165	24.0	264	0.3238	0.8785
0.112	25.0	275	0.3066	0.8889
0.1191	26.0	286	0.2427	0.9062
0.1293	27.0	297	0.2536	0.9201
0.2932	28.0	308	0.2707	0.8924
0.0918	29.0	319	0.2688	0.8924
0.1529	30.0	330	0.2715	0.8889
0.227	31.0	341	0.2664	0.9028
0.1044	32.0	352	0.2809	0.8993
0.0894	33.0	363	0.2863	0.8924
0.0566	34.0	374	0.2474	0.9201
0.0915	35.0	385	0.2428	0.9097
0.1136	36.0	396	0.2545	0.9097
0.0947	37.0	407	0.2599	0.9097
0.1012	38.0	418	0.2454	0.9167
0.0465	39.0	429	0.2435	0.9201
0.0299	40.0	440	0.2532	0.9062
0.0311	41.0	451	0.2298	0.9271
0.0796	42.0	462	0.2422	0.9167
0.058	43.0	473	0.2058	0.9306
0.0853	44.0	484	0.2266	0.9306
0.0868	45.0	495	0.2266	0.9236
0.0554	46.0	506	0.2163	0.9271
0.0508	47.0	517	0.2104	0.9306
0.0589	48.0	528	0.2172	0.9271
0.0369	49.0	539	0.2214	0.9271
0.0852	50.0	550	0.2241	0.9271

Framework versions

• Transformers 4.30.0.dev0
• Pytorch 1.12.1
• Datasets 2.12.0
• Tokenizers 0.13.1