metadata
license: mit
base_model: filevich/robertita-cased
tags:
- generated_from_trainer
datasets:
- fact2020
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: robertita-cased-finetuned-fact
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: fact2020
type: fact2020
config: fact2020
split: validation
args: fact2020
metrics:
- name: Precision
type: precision
value: 0.9946938891717209
- name: Recall
type: recall
value: 0.988775334205079
- name: F1
type: f1
value: 0.9893166033671081
- name: Accuracy
type: accuracy
value: 0.988775334205079
robertita-cased-finetuned-fact
This model is a fine-tuned version of filevich/robertita-cased on the fact2020 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0407
- Precision: 0.9947
- Recall: 0.9888
- F1: 0.9893
- Accuracy: 0.9888
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 |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 116 | 0.0554 | 0.9909 | 0.9817 | 0.9807 | 0.9817 |
| No log | 2.0 | 232 | 0.0415 | 0.9943 | 0.9879 | 0.9882 | 0.9879 |
| No log | 3.0 | 348 | 0.0407 | 0.9947 | 0.9888 | 0.9893 | 0.9888 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.4
- Tokenizers 0.13.3