tensorpool

Run your own code and manage clusters with Laravel's included tools. NoSQL and SQL supported.

435|12|Updated Jan 20, 2025
One-click install
npx skills add https://github.com/tensorpool/tensorpool --skill tensorpool
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tensorpool
Source: https://github.com/tensorpool/tensorpool/tree/main/skills/tensorpool
Command: npx skills add https://github.com/tensorpool/tensorpool --skill tensorpool

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps users migrate their local machine learning scripts to TensorPool GPU clusters using the interactive cluster workflow. Use this when you have a working local script and want to scale it up to professional GPU hardware.

Core Features & Use Cases

  • Analyze a local ML script to identify dependencies, data needs, and environment requirements.
  • Prepare the script for cloud execution by creating a requirements file, configuring environment variables, and testing a minimal run locally.
  • Provision and manage a GPU cluster, transfer code, set up the environment on the cluster, run the script, and retrieve results, with options for streaming logs and monitoring.
  • Use case: take a data science notebook or Python script and run it on H100/H200/B200/B300 GPUs with tp ssh for production-scale training or inference.

Quick Start

Prepare your local script for cloud execution by creating a minimal test run on a single GPU.

Frequently Asked Questions about tensorpool

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

FAQPage Schema
How do I migrate local ML scripts to a GPU cluster for production training?

To migrate local ML scripts to a GPU cluster, you can use an interactive workflow to analyze dependencies, provision TensorPool hardware like H100 or H200, transfer code via tp ssh, and execute runs remotely.

What is the best way to run a Python data science notebook on H100 GPUs?

Running a Python data science notebook on H100 GPUs involves preparing the local script environment, provisioning a scalable TensorPool cluster, transferring the code, and executing it remotely with streaming logs for monitoring.

Can I use an interactive cluster workflow to scale my machine learning script?

Yes, you can use an interactive cluster workflow to scale your machine learning script by analyzing it for environmental requirements, testing a minimal local run, provisioning cloud GPUs, and retrieving execution results.

Do I need to create a requirements file before moving my ML script to the cloud?

Yes, you need to create a requirements file and configure environment variables to prepare your ML script for cloud execution on GPU clusters before provisioning hardware and transferring your code.

How does code transfer and execution work when using TensorPool hardware?

Code transfer and execution on TensorPool hardware works by provisioning a GPU cluster, moving your local script to the cluster using tp ssh, setting up the environment, and running the script with options for streaming logs and result retrieval.

What steps are needed to prepare a local script for cloud execution on B200 or B300 GPUs?

Preparing a local script for cloud execution on B200 or B300 GPUs requires analyzing dependencies, configuring environment variables, testing a minimal local run, and provisioning the target hardware through the interactive cluster workflow.