Validated_Balanced_Raw_Data_model_boost7

This model is a fine-tuned version of facebook/convnext-base-224-22k on the Logiroad/Validated_Balanced_Raw_Dataset dataset. It achieves the following results on the evaluation set:

• Loss: 1.1008
• Accuracy: 0.5425

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: 1.5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 1337
• 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: 0.05
• num_epochs: 25.0
• label_smoothing_factor: 0.05

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	40	1.3316	0.3491
1.3694	2.0	80	1.2695	0.3726
1.291	3.0	120	1.2237	0.4151
1.2374	4.0	160	1.1886	0.4481
1.1815	5.0	200	1.1870	0.4387
1.1815	6.0	240	1.1726	0.4717
1.1479	7.0	280	1.1224	0.4858
1.0818	8.0	320	1.1309	0.4717
1.0507	9.0	360	1.1351	0.4811
1.0198	10.0	400	1.1314	0.5189
1.0198	11.0	440	1.1235	0.5047
1.0075	12.0	480	1.1136	0.5283
0.9692	13.0	520	1.1230	0.5094
0.919	14.0	560	1.1158	0.5
0.9306	15.0	600	1.1089	0.5236
0.9306	16.0	640	1.1008	0.5425
0.89	17.0	680	1.1071	0.5236
0.8853	18.0	720	1.1110	0.5236
0.8852	19.0	760	1.1026	0.5330
0.824	20.0	800	1.1056	0.5377
0.824	21.0	840	1.1088	0.5283
0.8327	22.0	880	1.1065	0.5283
0.832	23.0	920	1.1063	0.5330
0.8801	24.0	960	1.1065	0.5330
0.8372	25.0	1000	1.1065	0.5330

Framework versions

• Transformers 4.46.1
• Pytorch 2.3.0
• Datasets 3.1.0
• Tokenizers 0.20.3