mini-vanilla-target-tweet
This model is a fine-tuned version of google/bert_uncased_L-4_H-256_A-4 on the tweet_eval dataset. It achieves the following results on the evaluation set:
- Loss: 1.5603
- Accuracy: 0.7540
- F1: 0.7569
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 200
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
---|---|---|---|---|---|
0.9285 | 4.9 | 500 | 0.7493 | 0.7273 | 0.7207 |
0.4468 | 9.8 | 1000 | 0.7630 | 0.7460 | 0.7437 |
0.2194 | 14.71 | 1500 | 0.8997 | 0.7406 | 0.7455 |
0.1062 | 19.61 | 2000 | 1.0822 | 0.7433 | 0.7435 |
0.0568 | 24.51 | 2500 | 1.2225 | 0.7620 | 0.7622 |
0.0439 | 29.41 | 3000 | 1.3475 | 0.7513 | 0.7527 |
0.0304 | 34.31 | 3500 | 1.4999 | 0.7433 | 0.7399 |
0.0247 | 39.22 | 4000 | 1.5603 | 0.7540 | 0.7569 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1
- Datasets 2.7.1
- Tokenizers 0.13.2
- Downloads last month
- 12
Dataset used to train muhtasham/mini-vanilla-target-tweet
Evaluation results
- Accuracy on tweet_evalself-reported0.754
- F1 on tweet_evalself-reported0.757