metadata
license: apache-2.0
base_model: distilbert-base-uncased
tags:
- generated_from_trainer
datasets:
- emotion
metrics:
- accuracy
model-index:
- name: emotion_model
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: emotion
type: emotion
config: split
split: test
args: split
metrics:
- name: Accuracy
type: accuracy
value: 0.927
emotion_model
This model is a fine-tuned version of distilbert-base-uncased on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.3611
- Accuracy: 0.927
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: 0.0001
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.2619 | 1.0 | 250 | 0.2343 | 0.916 |
| 0.121 | 2.0 | 500 | 0.1432 | 0.93 |
| 0.1308 | 3.0 | 750 | 0.1565 | 0.9315 |
| 0.1012 | 4.0 | 1000 | 0.1595 | 0.925 |
| 0.0525 | 5.0 | 1250 | 0.1937 | 0.924 |
| 0.0635 | 6.0 | 1500 | 0.2635 | 0.9255 |
| 0.0183 | 7.0 | 1750 | 0.2726 | 0.9195 |
| 0.0156 | 8.0 | 2000 | 0.3324 | 0.9245 |
| 0.0036 | 9.0 | 2250 | 0.3614 | 0.925 |
| 0.011 | 10.0 | 2500 | 0.3611 | 0.927 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1