Yoda Ner

🚀 YODA: Google Feed Product Optimization Models

YODA consists of a series of models designed for optimizing Google Feed products. Our goal is to enhance the market reach of e - commerce by augmenting and improving specific metadata such as short titles, colors, measures, and more. Currently, over 300 companies with more than 3.5 million products are using YODA in production.

🚀 Quick Start

Prerequisites

• Flair (pip install flair)

Code Example

from flair.data import Sentence
from flair.models import SequenceTagger

# load tagger
tagger = SequenceTagger.load("lighthousefeed/yoda-ner")

# make example sentence
sentence = Sentence("Jean Paul Gaultier Classique - 50 ML Eau de Parfum  Damen Parfum.")

# predict NER tags
tagger.predict(sentence)

# print sentence
print(sentence)

# print predicted NER spans
print('The following NER tags are found:')

# iterate over entities and print
for entity in sentence.get_spans('ner'):
    print(entity)

✨ Features

NER for Product Feature Extraction

We've trained a NER model for extracting product features. It can retrieve data like colors, sizes, brands, and energy labels. Trained with over 3 million lines of product metadata, the model achieves the following scores:

Overall Results

• F - score (micro): 0.972
• F - score (macro): 0.9692
• Accuracy: 0.9461

Results by Class

Averaged Results

📄 License

This project is licensed under the ISC license.

📞 Contact

For any inquiries, please contact the lead ML developer Iván R. Gázquez. We're eager to hear about your use cases for this model!