dinov2-large-MEG7-10

This model is a fine-tuned version of facebook/dinov2-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3278
• Accuracy: 0.9324

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: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.354	0.9999	5312	0.4495	0.8662
0.2705	2.0	10625	0.3835	0.8900
0.2003	2.9999	15937	0.3546	0.9021
0.2712	4.0	21250	0.3297	0.9060
0.2028	4.9999	26562	0.3259	0.9099
0.1638	6.0	31875	0.3087	0.9183
0.1289	6.9999	37187	0.3260	0.9195
0.0752	8.0	42500	0.3197	0.9256
0.0643	8.9999	47812	0.3229	0.9276
0.0509	9.9991	53120	0.3278	0.9324

Framework versions

• Transformers 4.46.3
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.20.3