pollen-ner-450

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

• Loss: 0.3956
• Precision: 0.5539
• Recall: 0.7329
• F1: 0.6309

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	0.4861	0.4929	0.6305	0.5533
No log	2.0	114	0.4729	0.4844	0.6526	0.5560
No log	3.0	171	0.4552	0.4877	0.6767	0.5669
No log	4.0	228	0.4348	0.5103	0.6948	0.5884
No log	5.0	285	0.4222	0.5230	0.7088	0.6019
No log	6.0	342	0.4078	0.5356	0.7108	0.6109
No log	7.0	399	0.4068	0.5373	0.7229	0.6164
No log	8.0	456	0.4015	0.5486	0.7369	0.6290
0.7501	9.0	513	0.3970	0.5514	0.7329	0.6293
0.7501	10.0	570	0.3956	0.5539	0.7329	0.6309

Framework versions

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