distilbert-base-uncased-finetuned-emotion
This model is a fine-tuned version of distilbert-base-uncased on the emotion dataset for in the dataset in HG. It achieves the following results on the evaluation set:
- Loss: 0.2033
- Accuracy: 0.9275
- F1: 0.9273
Model description
This model is a copy of the model found in the book Natural Language Processing with Transformers.
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: 2
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
---|---|---|---|---|---|
0.806 | 1.0 | 250 | 0.2954 | 0.908 | 0.9062 |
0.2361 | 2.0 | 500 | 0.2033 | 0.9275 | 0.9273 |
Framework versions
- Transformers 4.13.0
- Pytorch 1.12.1+cu113
- Datasets 1.16.1
- Tokenizers 0.10.3
- Downloads last month
- 13
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Dataset used to train aambrioso/distilbert-base-uncased-finetuned-emotion
Evaluation results
- Accuracy on emotionself-reported0.927
- F1 on emotionself-reported0.927