Liang Claim this profile

Using Streamlit's built-in function to easily create a molecule grid viewer, and then you can assemble any content you want.

Implementing an Open Reaction Database downloader using Streamlit🎈.

No description.

This repository shows how to use the Hugging Face Library for molecule generation.

Unofficial mini implementation of 'Learning Continuous and Data-Driven Molecular Descriptors' paper. A concise codebase replicating the core architecture as a lightweight reference.

台灣獨立開發者專案列表：收錄台灣獨立開發者的網站、App、遊戲與開源作品

Parallel-universe redesigns of Taiwan .gov sites. First site: a HuggingFace × GitHub take on data.gov.tw + MoDA sovereign datasets.

This project uses a Variational Autoencoder (VAE) to generate SMILES strings for novel compound generation. The VAE model is trained on a dataset of existing chemical compounds and can generate new, valid SMILES strings, which may represent potentially new and useful chemical entities.

A lightweight interface for browsing public peptide datasets with chemical structures and drug-likeness metrics.

[IJCAI 2023 survey track]A curated list of resources for chemical pre-trained models

Compilation of chemoinformatics and machine learning techniques

新聞稿產生器

Splitaway TW — travel bill-splitting app prototypes (4 design directions) powered by real Taiwan tourism open data via Twinkle Hub. Built at a Code with Claude workshop. 台灣旅遊分帳原型 × 觀光開放資料。

No description.

Qwen3-TTS 台語（閩南語）批次音頻合成

CosyVoice3 Hokkien TTS setup with seed audio

PSMA-targeted peptidomimetic drug discovery: SELFIES-based Gemma-3-1b with GRPO docking reward

每日自動翻譯 HuggingFace Daily Papers 為繁體中文，含全文翻譯與圖表保留

FastAPI wrapper for Fun-CosyVoice3-0.5B: zero-shot voice cloning TTS with HuggingFace upload

No description.

A unified TypeScript SDK for building chat bots across Slack, Microsoft Teams, Google Chat, Discord, and more.

Dataset of PNG images from synthetically generated table layouts with annotations in JSONL files

No description.

Awesome Traditional Chinese Datasets

No description.

No description.

MolTrans: Molecular Interaction Transformer for Drug Target Interaction Prediction (Bioinformatics)

Taiwan law corpus, generated by https://github.com/ronnywang/twlaw

Implementation of the Paper "Learning Continuous and Data-Driven Molecular Descriptors by Translating Equivalent Chemical Representations" by Robin Winter, Floriane Montanari, Frank Noe and Djork-Arne Clevert.

ChemFold provides several methods for computing train-validation-test splits, both in federated and non-federated (ordinary) ML settings.

No description.

No description.

No description.

RDKit related blog posts, notebooks, and data.

No description.

SMILES Pair Encoding: A data-driven substructure representation of chemicals

Automatic prototype based drug generation