CNEC_1_1_ext_robeczech-base

This model is a fine-tuned version of ufal/robeczech-base on the cnec dataset. It achieves the following results on the evaluation set:

• Loss: 0.1985
• Precision: 0.8639
• Recall: 0.8990
• F1: 0.8811
• Accuracy: 0.9633

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.2585	6.85	1000	0.1912	0.8276	0.8696	0.8481	0.9550
0.1224	13.7	2000	0.1807	0.8455	0.8894	0.8669	0.9586
0.0788	20.55	3000	0.1715	0.8624	0.8974	0.8795	0.9643
0.0562	27.4	4000	0.1782	0.8650	0.9043	0.8842	0.9633
0.0432	34.25	5000	0.1856	0.8598	0.9017	0.8803	0.9640
0.0346	41.1	6000	0.1975	0.8622	0.8963	0.8789	0.9630
0.0306	47.95	7000	0.1985	0.8639	0.8990	0.8811	0.9633

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0