metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- amazon_reviews_multi
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: electra-small-finetuned-amazon-review
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: amazon_reviews_multi
type: amazon_reviews_multi
args: es
metrics:
- name: Accuracy
type: accuracy
value: 0.4948
- name: F1
type: f1
value: 0.49332463542809535
- name: Precision
type: precision
value: 0.4921725374649701
- name: Recall
type: recall
value: 0.4948
electra-small-finetuned-amazon-review
This model is a fine-tuned version of google/electra-small-discriminator on the amazon_reviews_multi dataset. It achieves the following results on the evaluation set:
- Loss: 1.1647
- Accuracy: 0.4948
- F1: 0.4933
- Precision: 0.4922
- Recall: 0.4948
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: 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
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
---|---|---|---|---|---|---|---|
1.4061 | 1.0 | 1000 | 1.2279 | 0.4496 | 0.4230 | 0.4359 | 0.4496 |
1.1941 | 2.0 | 2000 | 1.1783 | 0.4782 | 0.4586 | 0.4567 | 0.4782 |
1.0997 | 3.0 | 3000 | 1.1648 | 0.4966 | 0.4785 | 0.4805 | 0.4966 |
1.0265 | 4.0 | 4000 | 1.1507 | 0.4996 | 0.4932 | 0.4920 | 0.4996 |
0.9736 | 5.0 | 5000 | 1.1647 | 0.4948 | 0.4933 | 0.4922 | 0.4948 |
Framework versions
- Transformers 4.15.0
- Pytorch 1.10.0+cu111
- Datasets 1.17.0
- Tokenizers 0.10.3