unimol

Official

Embed molecules and train ML models with Uni-Mol.

Authorjinzhezenggroup
Version1.0.0
Installs0

System Documentation

What problem does it solve?

UniMol-based workflows often require chaining representation extraction, model training, and prediction in a reproducible CLI workflow with validation for SMILES. This skill provides a minimal, standardized wrapper to run unimol-tools across embedding, training, and prediction steps, ensuring consistent outputs and error handling.

Core Features & Use Cases

  • Embedding extraction: generate molecular representations (embeddings) from SMILES datasets.
  • Model training: support regression, classification, and multilabel tasks with automatic target handling on CSV/SMI inputs.
  • Prediction: perform property predictions on new data and output a CSV with predictions and logs for invalid entries.
  • Robustness: environment detection, optional GPU usage, and RDKit-based SMILES validation to prevent crashes.

Quick Start

Run uv run python <skill_path>/scripts/unimol_helper.py repr/train/predict to perform the corresponding task.

Dependency Matrix

Required Modules

unimol-toolstorchrdkitpandasnumpy

Components

scripts

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: unimol
Download link: https://github.com/jinzhezenggroup/computational-chemistry-agent-skills/archive/main.zip#unimol

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.