metadata
language:
- en
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- tweet_eval
metrics:
- accuracy
model-index:
- name: distilbert-base-cased-finetuned-tweeteval
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: tweet_eval
type: tweet_eval
config: emotion
split: validation
args: emotion
metrics:
- name: Accuracy
type: accuracy
value: 0.7887700534759359
distilbert-base-cased-finetuned-tweeteval
This model is a fine-tuned version of distilbert-base-cased on the tweet_eval dataset. It achieves the following results on the evaluation set:
- Loss: 0.7720
- Accuracy: 0.7888
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 204 | 0.6867 | 0.7647 |
| No log | 2.0 | 408 | 0.6318 | 0.7968 |
| 0.6397 | 3.0 | 612 | 0.6931 | 0.7834 |
| 0.6397 | 4.0 | 816 | 0.7631 | 0.7754 |
| 0.2064 | 5.0 | 1020 | 0.7720 | 0.7888 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3