README.md

metadata

license: mit
base_model: nlptown/bert-base-multilingual-uncased-sentiment
tags:
  - generated_from_trainer
datasets:
  - amazon_reviews_multi
metrics:
  - accuracy
  - f1
model-index:
  - name: amazon-reviews-sentiment-bert-base-uncased-6000-samples
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: amazon_reviews_multi
          type: amazon_reviews_multi
          config: en
          split: validation
          args: en
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7678571428571429
          - name: F1
            type: f1
            value: 0.7167992873886065

amazon-reviews-sentiment-bert-base-uncased-6000-samples

This model is a fine-tuned version of nlptown/bert-base-multilingual-uncased-sentiment on the amazon_reviews_multi dataset. It achieves the following results on the evaluation set:

• Loss: 0.5890
• Accuracy: 0.7679
• F1: 0.7168

Predicted labels

• LABEL_0: Negative review
• LABEL_1: Neutral review
• LABEL_2: Positive review

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: 32
• eval_batch_size: 32
• 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
No log	1.0	188	0.5745	0.7586	0.7149
No log	2.0	376	0.5890	0.7679	0.7168

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.0
• Datasets 2.14.6.dev0
• Tokenizers 0.13.3