pollen-ner2-450

This model is a fine-tuned version of DeepPavlov/bert-base-bg-cs-pl-ru-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4605
• Precision: 0.4883
• Recall: 0.6305
• F1: 0.5504

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• 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: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1
No log	1.0	57	0.5754	0.4231	0.5361	0.4730
No log	2.0	114	0.5404	0.4237	0.5462	0.4772
No log	3.0	171	0.5230	0.4407	0.5743	0.4987
No log	4.0	228	0.5053	0.4470	0.5843	0.5065
No log	5.0	285	0.4844	0.4619	0.5843	0.5160
No log	6.0	342	0.4810	0.4708	0.6145	0.5331
No log	7.0	399	0.4710	0.4784	0.6225	0.5410
No log	8.0	456	0.4631	0.4822	0.6245	0.5442
0.9019	9.0	513	0.4615	0.4852	0.6265	0.5469
0.9019	10.0	570	0.4605	0.4883	0.6305	0.5504

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.7.0+cu128
• Datasets 3.5.0
• Tokenizers 0.21.1