kpi-priority-model

This model is a fine-tuned version of answerdotai/ModernBERT-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7414

• Accuracy: 0.7288

• F1: 0.7266

• Classification Report: precision recall f1-score support

     Low       0.68      0.83      0.75        94
  Medium       0.66      0.52      0.58       111
    High       0.69      0.73      0.71       202
  

  Critical 0.81 0.78 0.80 253

  accuracy 0.73 660 macro avg 0.71 0.72 0.71 660

weighted avg 0.73 0.73 0.73 660

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• 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: 100
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Classification Report
2.3852	1.0	165	0.8176	0.6833	0.6838	precision recall f1-score support

     Low       0.65      0.85      0.74        94
  Medium       0.55      0.55      0.55       111
    High       0.62      0.78      0.69       202
Critical       0.88      0.60      0.72       253

accuracy                           0.68       660

macro avg 0.68 0.70 0.67 660 weighted avg 0.71 0.68 0.68 660 | | 1.3946 | 2.0 | 330 | 0.7414 | 0.7288 | 0.7266 | precision recall f1-score support

     Low       0.68      0.83      0.75        94
  Medium       0.66      0.52      0.58       111
    High       0.69      0.73      0.71       202
Critical       0.81      0.78      0.80       253

accuracy                           0.73       660

macro avg 0.71 0.72 0.71 660 weighted avg 0.73 0.73 0.73 660 |

Framework versions

• Transformers 4.49.0.dev0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0