skills.rest
dawsonblock
dawsonblockCOMMUNITY
Explore all skills by this author →
GitHub

cli-anything-unimol-tools

Train and infer molecular property prediction models from SMILES CSV datasets via Uni-Mol Tools CLI.

Updated May 7, 2026
Search Tags:#cli#training#inference#machine-learning#smiles#molecular-property#unimol
One-click install
npx skills add https://github.com/dawsonblock/JUDGE_ATLAS --skill cli-anything-unimol-tools-dawsonblock
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cli-anything-unimol-tools
Source: https://github.com/dawsonblock/JUDGE_ATLAS/tree/main/external/CLI-Anything-main/skills/cli-anything-unimol-tools
Command: npx skills add https://github.com/dawsonblock/JUDGE_ATLAS --skill cli-anything-unimol-tools-dawsonblock

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It streamlines Uni-Mol based molecular property modeling so researchers can go from CSV datasets to trained models and predictions without stitching together multiple ad-hoc scripts.

Core Features & Use Cases

  • Project management: create, list, and switch named experiment projects so runs stay organized.
  • End-to-end modeling for 5 task types: handle classification, regression, multiclass, multilabel classification, and multilabel regression.
  • Model lifecycle management: list models, inspect performance, rank candidates, and track historical results for iteration.
  • Storage analysis and cleanup: measure disk usage and remove underperforming models automatically or by criteria.
  • Automation via JSON mode: run commands non-interactively and get machine-readable output.

Quick Start

Train a regression model by running: train --data-path train.csv --target-col affinity --task-type regression --epochs 10.

Frequently Asked Questions about cli-anything-unimol-tools

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I train a molecular property prediction model from a CSV file?▼

To train a molecular property prediction model, run the command-line interface with your CSV dataset, specifying the target column, task type, and epochs. The system handles single-label, multiclass, and multilabel variants automatically.

Can I automate molecular property inference and get machine-readable results?▼

Yes, you can automate molecular property inference by using the JSON mode flag to run commands non-interactively, which returns deterministic machine-readable JSON results suitable for automation pipelines.

What machine learning task types are supported for molecular property modeling?▼

Molecular property modeling supports five task types: classification, regression, multiclass classification, multilabel classification, and multilabel regression, covering a wide range of predictive workflows for SMILES datasets.

How do I manage disk space when storing multiple trained Uni-Mol models?▼

You can manage disk space by analyzing storage usage and removing underperforming models automatically or by specific criteria, ensuring your experiment projects stay organized without consuming excessive disk space.

Does the Uni-Mol Tools CLI require specific input formats for SMILES datasets?▼

The Uni-Mol Tools CLI requires a CSV input format containing a SMILES column and one or more target columns. It does not list external dependencies, operating as a standalone command-line interface for your datasets.

Related Skills

Explore More Education & Research Skills→
jinzhezenggroupjinzhezenggroup
124

rdkit-repr

Compute RDKit molecular descriptors and fingerprints from SMILES data.

Official
Intermediate
grahama1970grahama1970
1

treesitter

Parse source files to extract functions, classes, and symbols into structured JSON.

Community
Intermediate
qcmuuqcmuu

weights-and-biases

Track experiment runs, log metrics, and manage hyperparameters with Weights & Biases.

Community
Advanced
HKUDSHKUDS
46.8k

cli-anything-unimol-tools

Manage molecular property prediction models via command-line interface.

Official
Intermediate