pollen-ner-500

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: 0.3313
• Precision: 0.6161
• Recall: 0.7992
• F1: 0.6958

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	63	0.3821	0.5504	0.7450	0.6331
No log	2.0	126	0.3790	0.5526	0.7590	0.6396
No log	3.0	189	0.3655	0.5687	0.7651	0.6524
No log	4.0	252	0.3649	0.5714	0.7791	0.6593
No log	5.0	315	0.3379	0.6071	0.7851	0.6848
No log	6.0	378	0.3412	0.6015	0.7912	0.6834
No log	7.0	441	0.3354	0.6071	0.7912	0.6870
0.6175	8.0	504	0.3355	0.6070	0.7972	0.6892
0.6175	9.0	567	0.3321	0.6142	0.7992	0.6946
0.6175	10.0	630	0.3313	0.6161	0.7992	0.6958

Framework versions

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