SentiChain

About Us

SentiChain is a blockchain-based sentiment analysis project that demonstrates how sentiment data can be anchored on-chain and refined into insights. We develop innovative solutions for text vectorization and embedding processing, making complex text analysis more accessible and efficient.

Key Products

• Aparecium: A state-of-the-art Python package for revealing text from embedding vectors, particularly designed to work with SentiChain embeddings. Named after the Harry Potter spell that reveals hidden writing, Aparecium provides tools for converting between text and vector representations, as well as reversing the embedding process to recover original text.

Features

• Text Vectorization using pre-trained transformer models
• Embedding Reversal with Transformer-based sequence-to-sequence architecture
• Seamless Integration with SentiChain embeddings
• Modern Architecture built on PyTorch and Transformers
• Extensible Design for custom model integration

Technology Stack

• Python
• PyTorch
• Transformers
• OpenAI Integration
• Data management

Open Source

We believe in the power of open source and actively contribute to the developer community. Our projects are available under the MIT License.

Contact

• GitHub: SentiChain
• Email: [email protected]

License

All our projects are licensed under the MIT License, promoting open collaboration and innovation in the developer community.