Binary sentiment classifier for Persian, fine-tuned from
shekar-ai/albert-base-v2-persian-zwnj-naab-mlm
on the Snappfood corpus of Persian food-delivery reviews.
1from shekar import SentimentClassifier
2
3classifier = SentimentClassifier()
4print(classifier("غذا عالی بود و خیلی سریع رسید."))
5print(classifier("کیفیت غذا افتضاح بود و سرد به دستم رسید."))
1('positive', 0.9934062957763672)
2('negative', 0.9793806672096252)
1from transformers import pipeline
2
3classifier = pipeline(
4 "text-classification",
5 model="shekar-ai/albert-base-v2-persian-sentiment-snapfood",
6)
7
8for prediction in classifier([
9 "غذا عالی بود و خیلی سریع رسید.",
10 "کیفیت غذا افتضاح بود و سرد به دستم رسید.",
11]):
12 print(prediction["label"], round(prediction["score"], 4))
1POSITIVE 0.9957
2NEGATIVE 0.9768
1@article{Amirivojdan2025Shekar,
2 author = {Amirivojdan, Ahmad},
3 title = {{Shekar: A Python Toolkit for Persian Natural Language Processing}},
4 journal = {Journal of Open Source Software},
5 volume = {10},
6 number = {114},
7 pages = {9128},
8 year = {2025},
9 doi = {10.21105/joss.09128},
10 url = {https://joss.theoj.org/papers/10.21105/joss.09128}
11}
