dnzblgn's picture
Update README.md
095e140 verified
metadata
license: apache-2.0
datasets:
  - McAuley-Lab/Amazon-Reviews-2023
language:
  - en
metrics:
  - accuracy
base_model:
  - microsoft/deberta-v3-base
pipeline_tag: text-classification
widget:
  - text: The product didn't arrive on time and was damaged.
library_name: transformers
safetensors: true

sentiment_mapping = {1: "Negative", 0: "Positive"}

Training Details

The model was trained on the McAuley-Lab/Amazon-Reviews-2023 dataset. This dataset contains labeled customer reviews from Amazon, focusing on two primary categories: Positive and Negative.

Training Hyperparameters

  • Model: microsoft/deberta-v3-base
  • Learning Rate: 3e-5
  • Epochs: 6
  • Train Batch Size: 16
  • Gradient Accumulation Steps: 2
  • Weight Decay: 0.015
  • Warm-up Ratio: 0.1

Evaluation

The model was evaluated using a subset of the Amazon reviews dataset, focusing on the binary classification of text as either positive or negative.

Metrics

Accuracy: 0.98

Precision: 0.98

Recall: 0.99

F1-Score: 0.98

from transformers import pipeline

classifier = pipeline("text-classification", model="dnzblgn/Sentiment-Analysis-Customer-Reviews")
result = classifier("The product didn't arrive on time and was damaged.")
print(result)