gs-aristoBERTo

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

• Loss: 1.9884
• Top1: 20.0
• Top5: 34.0
• Top10: 43.6667
• Top20: 47.0
• Bertscore F1 Top1: 81.5387
• Bertscore F1 Top1 Mean: 81.5387
• Bertscore F1 Top5: 86.5058
• Bertscore F1 Top5 Mean: 79.3804
• Bertscore F1 Top10: 88.8704
• Bertscore F1 Top10 Mean: 78.4608
• Bertscore F1 Top20: 89.9777
• Bertscore F1 Top20 Mean: 77.3623
• Cos Sim Top1 Max: 67.2364
• Cos Sim Top1 Mean: 67.2364
• Cos Sim Top5 Max: 77.8154
• Cos Sim Top5 Mean: 63.4673
• Cos Sim Top10 Max: 82.1975
• Cos Sim Top10 Mean: 62.2533
• Cos Sim Top20 Max: 84.7292
• Cos Sim Top20 Mean: 61.5174
• Composite Score: 43.6182

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: 1.187383488773285e-06
• train_batch_size: 128
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 0.1
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Top1	Top5	Top10	Top20	Bertscore F1 Top1	Bertscore F1 Top1 Mean	Bertscore F1 Top5	Bertscore F1 Top5 Mean	Bertscore F1 Top10	Bertscore F1 Top10 Mean	Bertscore F1 Top20	Bertscore F1 Top20 Mean	Cos Sim Top1 Max	Cos Sim Top1 Mean	Cos Sim Top5 Max	Cos Sim Top5 Mean	Cos Sim Top10 Max	Cos Sim Top10 Mean	Cos Sim Top20 Max	Cos Sim Top20 Mean	Composite Score
2.2020	1.0	192	2.0431	19.6667	34.6667	44.0	47.0	81.5457	81.5457	86.5778	79.3828	88.8950	78.4672	89.9488	77.3708	67.1874	67.1874	77.9751	63.6169	82.4312	62.3734	84.8573	61.6775	43.4271
2.1106	2.0	384	1.9966	20.0	34.0	43.6667	47.0	81.5387	81.5387	86.5058	79.3804	88.8704	78.4608	89.9777	77.3623	67.2364	67.2364	77.8154	63.4673	82.1975	62.2533	84.7292	61.5174	43.6182

Framework versions

• Transformers 5.9.0
• Pytorch 2.12.0+cu126
• Datasets 3.6.0
• Tokenizers 0.22.2