Skip to content

PerSent v1.3.2 Release Note

Latest

Choose a tag to compare

@RezaGooner RezaGooner released this 04 Aug 12:13
· 20 commits to main since this release

🚀 What’s new in this release?

✅ Pre-trained Model Now Included!

  • No more manual downloads: PerSent now comes bundled with the pre-trained sentiment analysis model – it is installed automatically with the library.
  • Plug-and-play: Just install the PerSent package via pip, and you are ready to analyze text without any extra setup or downloading files.
  • Internal loading: The analyzer will automatically find and use the embedded model asset, so you no longer need to specify a separate model path.

♻️ Changelog (compared to previous release):

  • 🔗 Packaged the pre-trained model file as a Python package asset
    The weighted_sentiment_model.joblib is now included in the package and will be installed automatically.
  • 🔨 Simplified installation and setup
    You no longer need to download or manually copy any model files.
  • 🧽 Improved reliability of model loading
    The library will always find the correct embedded model after installation, even inside virtual environments.
  • 📝 [Docs] Updated installation and usage instructions in the README to reflect the zero-setup workflow.

🛠 How to upgrade

pip install -U persent

⚡️ Getting started

Minimal example (after installation):

from persent import SentimentAnalyzer

analyzer = SentimentAnalyzer()
analyzer.loadModel()      # No argument needed default – bundled model is loaded automatically!
result = analyzer.analyzeText("از دستت ناراحتم")

print(result)

💡 Feedback & Contributions

If you find bugs or want to contribute, please feel free to open an issue or pull request.

Special thanks to all early users and contributors!

RezaGooner

Full Changelog: V1.2.0...V1.3.2