metadata
license: apache-2.0
base_model: distilbert-base-cased
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: distilbert-cased-reviews-v1
results: []
widget:
- text: >-
Red Hot Chili Peppers on vinyl has been a disappointing experience.. I had
to return both “By The Way” and “Stadium Arcadium” because there were
skips on almost all of it.. Kind of made it seem like the record label
just went cheap, which is a disservice to anyone that actually listens to
their vinyl...This “Greatest Hits” compilation did not have the same
problems as the other two I bought. It sounded as it should have, and
there were no skips.
datasets:
- yyu/amazon-attrprompt
language:
- en
distilbert-cased-reviews-v1
This model is a fine-tuned version of distilbert-base-cased on yyu/amazon-attrprompt dataset. It achieves the following results on the evaluation set:
- Loss: 1.9022
- Accuracy: {'accuracy': 0.7478260869565218}
- F1 Score: {'f1': 0.7350319489971969}
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Score |
|---|---|---|---|---|---|
| 0.2514 | 1.0 | 1728 | 1.8422 | {'accuracy': 0.7347826086956522} | {'f1': 0.7217427746565059} |
| 0.2709 | 2.0 | 3456 | 1.8755 | {'accuracy': 0.7347826086956522} | {'f1': 0.7200496580345588} |
| 0.0912 | 3.0 | 5184 | 1.9022 | {'accuracy': 0.7478260869565218} | {'f1': 0.7350319489971969} |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0