test-trainer

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

• Loss: 0.7291
• Accuracy: 0.9107
• F1: 0.9122
• Precision: 0.9172
• Recall: 0.9107

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: 2e-05
• train_batch_size: 10
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	1.0	50	1.6720	0.4821	0.4134	0.3870	0.4821
No log	2.0	100	1.4652	0.6429	0.6126	0.7414	0.6429
No log	3.0	150	1.1742	0.7321	0.7210	0.7792	0.7321
No log	4.0	200	0.9813	0.8393	0.8433	0.8589	0.8393
No log	5.0	250	0.8312	0.8214	0.8164	0.8516	0.8214
No log	6.0	300	0.7291	0.9107	0.9122	0.9172	0.9107

Framework versions

• Transformers 4.46.2
• Pytorch 2.2.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3