README.md

metadata

tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: rubert-base-cased-token
    results: []
widget:
  - text: >-
      Однажды я посетил прекрасный городок в горах. На его улицах росли
      удивительные цветы.
    example_title: Example_1

rubert-base-cased-token

This model is a fine-tuned version of DeepPavlov/rubert-base-cased on the OpenCorpora dataset opencorpora.org. It achieves the following results on the evaluation set:

• Loss: 0.2595
• Precision: 0.9304
• Recall: 0.9334
• F1: 0.9319
• Accuracy: 0.9424

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

Tokens classification from OpenCorpora: opencorpora.org

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	69	0.6926	0.7731	0.7674	0.7702	0.8200
No log	2.0	138	0.3744	0.8665	0.8807	0.8735	0.9003
No log	3.0	207	0.2891	0.9004	0.9071	0.9037	0.9231
No log	4.0	276	0.2566	0.9123	0.9217	0.9170	0.9327
No log	5.0	345	0.2587	0.9211	0.9255	0.9233	0.9366
No log	6.0	414	0.2472	0.9264	0.9289	0.9276	0.9401
No log	7.0	483	0.2589	0.9267	0.9313	0.9290	0.9406
0.3825	8.0	552	0.2559	0.9286	0.9334	0.9310	0.9416
0.3825	9.0	621	0.2578	0.9304	0.9339	0.9321	0.9425
0.3825	10.0	690	0.2595	0.9304	0.9334	0.9319	0.9424

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.0+cu116
• Datasets 2.8.0
• Tokenizers 0.13.2