cifar100-ButterflyMlpNet

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 9.0995
• Accuracy: 0.3216

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: 0.001
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.4676	1.0	3125	3.5962	0.1578
3.3095	2.0	6250	3.3498	0.1981
2.9699	3.0	9375	3.0833	0.2476
2.8611	4.0	12500	2.9639	0.2718
2.6343	5.0	15625	2.8729	0.3003
2.4611	6.0	18750	2.8173	0.3188
2.1236	7.0	21875	2.8665	0.3168
1.8903	8.0	25000	3.0810	0.3274
1.5247	9.0	28125	3.3761	0.3209
1.0512	10.0	31250	3.7849	0.3207
0.8395	11.0	34375	4.2214	0.3155
0.4714	12.0	37500	4.8652	0.3097
0.3216	13.0	40625	5.6464	0.3134
0.1419	14.0	43750	6.5424	0.3151
0.0718	15.0	46875	7.3355	0.3173
0.0472	16.0	50000	8.1995	0.3225
0.0058	17.0	53125	8.6810	0.3211
0.0033	18.0	56250	8.9136	0.3202
0.0069	19.0	59375	9.0834	0.3214
0.0008	20.0	62500	9.0995	0.3216

Framework versions

• Transformers 4.55.2
• Pytorch 2.7.1+cu126
• Datasets 4.0.0
• Tokenizers 0.21.4