metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- tweet_eval
metrics:
- f1
model-index:
- name: irony_trained
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: tweet_eval
type: tweet_eval
args: irony
metrics:
- name: F1
type: f1
value: 0.6851011633121422
irony_trained
This model is a fine-tuned version of distilbert-base-uncased on the tweet_eval dataset. It achieves the following results on the evaluation set:
- Loss: 1.6471
- F1: 0.6851
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: 2.6774391860025942e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 0
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.6589 | 1.0 | 716 | 0.6187 | 0.6646 |
| 0.5494 | 2.0 | 1432 | 0.9314 | 0.6793 |
| 0.3369 | 3.0 | 2148 | 1.3468 | 0.6833 |
| 0.2129 | 4.0 | 2864 | 1.6471 | 0.6851 |
Framework versions
- Transformers 4.12.5
- Pytorch 1.9.1
- Datasets 1.16.1
- Tokenizers 0.10.3