Deploy and run ML experiments on local or remote GPU servers.
npx skills add https://github.com/THUFanZd/Sewed_pipeline --skill run-experiment-thufanzd
Please help me install this Agent Skill. Skill: run-experiment Source: https://github.com/THUFanZd/Sewed_pipeline/tree/main/.agents/skills/run-experiment Command: npx skills add https://github.com/THUFanZd/Sewed_pipeline --skill run-experiment-thufanzd
Deploy and run ML experiments on local or remote GPU environments, unifying setup, execution, and verification to reduce manual orchestration and downtime.
Run an ML experiment on the target GPU server using the provided script and arguments.
High-intent search queries and answers about installing and using this skill.
To run ML experiments on a remote GPU server, this solution orchestrates the deployment by syncing code changes, validating CUDA driver readiness, and launching training jobs in isolated screen sessions with specific GPU binding. It relies on AGENTS.md configuration to determine server paths and conda environments.
Yes, you can use W&B integration for experiment tracking when launching training jobs remotely. By enabling Weights & Biases in the AGENTS.md configuration, the execution workflow automatically logs metrics and supports experiment reproducibility throughout the remote training process.
You do not need to manually sync code changes to the GPU server before launching training. The deployment workflow automatically synchronizes code changes using rsync by default or git-based methods, specifically avoiding large data transfers to optimize the remote setup process.
The system validates GPU readiness by performing pre-flight checks that confirm available GPUs, verify CUDA and driver readiness, and ensure compatibility before launching training. This validation prevents execution errors and downtime on the target remote or local server.
The best way to run ML experiments on either environment is through environment-driven deployment. The system detects whether the target is a local machine or remote server using project metadata in AGENTS.md, automatically applying the correct workflow, GPU binding, and conda environment isolation for each scenario.
kitcafDeploy and run ML experiments on local or remote GPU servers.
gztcodeAutomate ML experiment tracking and collaboration with Weights & Biases.
goupup-aiAutomate ML training experiment deployment across local, remote, Vast.ai, and Modal GPU environments.
pluginagentmarketplaceAutomate Linux data science environment setup with GPU drivers and ML frameworks.
