metadata
license: mit
base_model: digitalepidemiologylab/covid-twitter-bert-v2
tags:
- generated_from_trainer
metrics:
- f1
- recall
- precision
model-index:
- name: covid_bert-e3-b16-v2-w0.01-dev1
results: []
covid_bert-e3-b16-v2-w0.01-dev1
This model is a fine-tuned version of digitalepidemiologylab/covid-twitter-bert-v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9608
- F1: 0.7913
- Recall: 0.7913
- Precision: 0.7913
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: 4
- eval_batch_size: 4
- 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 | F1 | Recall | Precision |
|---|---|---|---|---|---|---|
| 0.8059 | 1.0 | 753 | 0.6912 | 0.7558 | 0.7558 | 0.7558 |
| 0.5563 | 2.0 | 1506 | 0.7618 | 0.7804 | 0.7804 | 0.7804 |
| 0.3678 | 3.0 | 2259 | 0.9608 | 0.7913 | 0.7913 | 0.7913 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3