metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- collection3
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: rubert-finetuned-collection3
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: collection3
type: collection3
config: default
split: train
args: default
metrics:
- name: Precision
type: precision
value: 0.9354685646500593
- name: Recall
type: recall
value: 0.9577362156910372
- name: F1
type: f1
value: 0.9464714354296688
- name: Accuracy
type: accuracy
value: 0.986481047855993
rubert-finetuned-collection3
This model is a fine-tuned version of sberbank-ai/ruBert-base on the collection3 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0514
- Precision: 0.9355
- Recall: 0.9577
- F1: 0.9465
- Accuracy: 0.9865
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.0794 | 1.0 | 1163 | 0.0536 | 0.9178 | 0.9466 | 0.9320 | 0.9825 |
0.0391 | 2.0 | 2326 | 0.0512 | 0.9228 | 0.9553 | 0.9388 | 0.9853 |
0.0191 | 3.0 | 3489 | 0.0514 | 0.9355 | 0.9577 | 0.9465 | 0.9865 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0.dev20220929+cu117
- Datasets 2.8.0
- Tokenizers 0.13.2