README.md

metadata

license: apache-2.0
tags:
  - image-classification
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: croupier-creature-classifier
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: croupier-mtg-dataset
          type: imagefolder
          config: alkzar90--croupier-mtg-dataset
          split: train
          args: alkzar90--croupier-mtg-dataset
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7471264367816092

croupier-creature-classifier

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

• Loss: 0.8477
• Accuracy: 0.7471

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: 3e-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: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1159	1.1	100	1.1144	0.6118
0.8183	2.2	200	0.9109	0.6882
0.6829	3.3	300	0.7677	0.7235
0.5575	4.4	400	0.7670	0.6765
0.4644	5.49	500	0.8460	0.6647
0.3096	6.59	600	0.7082	0.7529
0.305	7.69	700	0.6939	0.7647
0.3349	8.79	800	0.7285	0.7235
0.36	9.89	900	0.7664	0.7294
0.3184	10.99	1000	0.6807	0.7588
0.2815	12.09	1100	0.7408	0.7353
0.1745	13.19	1200	0.7528	0.7294
0.1894	14.29	1300	0.7634	0.7471
0.1641	15.38	1400	0.7209	0.7647
0.1932	16.48	1500	0.9091	0.7
0.1609	17.58	1600	0.7208	0.7588
0.132	18.68	1700	0.8487	0.7588
0.1903	19.78	1800	0.7912	0.7471
0.121	20.88	1900	0.6735	0.7471
0.1903	21.98	2000	0.6692	0.7824
0.176	23.08	2100	0.8351	0.7176
0.1186	24.18	2200	0.7318	0.7471
0.1424	25.27	2300	0.7860	0.7588
0.144	26.37	2400	0.7021	0.7882
0.1088	27.47	2500	0.8109	0.7471
0.1019	28.57	2600	0.8157	0.7471
0.0947	29.67	2700	0.8028	0.7588
0.1715	30.77	2800	0.8345	0.7471
0.1046	31.87	2900	0.8578	0.7412
0.1367	32.97	3000	0.7670	0.7882
0.1339	34.07	3100	0.7763	0.7647
0.1194	35.16	3200	0.7727	0.7706
0.151	36.26	3300	0.8272	0.7471
0.0646	37.36	3400	0.7721	0.7765
0.0801	38.46	3500	0.8171	0.7529
0.1038	39.56	3600	0.9464	0.7059
0.16	40.66	3700	0.8005	0.7706
0.1151	41.76	3800	0.8784	0.7471
0.1159	42.86	3900	0.8598	0.7471
0.0575	43.96	4000	0.8543	0.7529
0.164	45.05	4100	0.8659	0.7588
0.1319	46.15	4200	0.8854	0.7412
0.0489	47.25	4300	0.7508	0.7588
0.0678	48.35	4400	0.8784	0.7353
0.0832	49.45	4500	0.7248	0.7765

Framework versions

• Transformers 4.21.1
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1