metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- health_fact
metrics:
- accuracy
- f1
model-index:
- name: distilbert-base-uncased-finetuned-health_facts
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: health_fact
type: health_fact
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.6186161449752883
- name: F1
type: f1
value: 0.6389624543035197
distilbert-base-uncased-finetuned-health_facts
This model is a fine-tuned version of distilbert-base-uncased on the health_fact dataset. It achieves the following results on the evaluation set:
- Loss: 1.4752
- Accuracy: 0.6186
- F1: 0.6390
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: 64
- eval_batch_size: 64
- 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 | Accuracy | F1 |
---|---|---|---|---|---|
1.1402 | 1.0 | 154 | 0.9476 | 0.5585 | 0.6082 |
0.9477 | 2.0 | 308 | 0.9208 | 0.5717 | 0.6164 |
0.8218 | 3.0 | 462 | 0.9474 | 0.5741 | 0.6190 |
0.7151 | 4.0 | 616 | 1.0258 | 0.5807 | 0.6213 |
0.6139 | 5.0 | 770 | 1.2066 | 0.6301 | 0.6577 |
0.5482 | 6.0 | 924 | 1.3031 | 0.6269 | 0.6542 |
0.4769 | 7.0 | 1078 | 1.3301 | 0.5890 | 0.6204 |
0.4337 | 8.0 | 1232 | 1.3727 | 0.6096 | 0.6326 |
0.3891 | 9.0 | 1386 | 1.4251 | 0.6104 | 0.6344 |
0.358 | 10.0 | 1540 | 1.4752 | 0.6186 | 0.6390 |
Framework versions
- Transformers 4.11.3
- Pytorch 1.10.0
- Datasets 1.16.1
- Tokenizers 0.10.3