Automate iterative ML research loops by modifying training code and evaluating metrics.
npx skills add https://github.com/aviskaar/open-org --skill autoresearch-aviskaar
Please help me install this Agent Skill. Skill: autoresearch Source: https://github.com/aviskaar/open-org/tree/main/skills/autoresearch Command: npx skills add https://github.com/aviskaar/open-org --skill autoresearch-aviskaar
This Skill automates the process of iterating on machine learning model training, allowing for continuous improvement without manual intervention.
Use the autoresearch skill to autonomously run an ML research loop on train.py with a 5-minute time budget per experiment, optimizing for validation accuracy.
High-intent search queries and answers about installing and using this skill.
Automating an iterative ML research loop involves automatically modifying training code, executing timed experiments, and evaluating a single metric to retain beneficial changes autonomously. You configure training script paths, time budgets, and optimization metrics to let the agent discover improved configurations.
Yes, autonomous hyperparameter tuning supports cross-platform execution across CUDA, Apple Silicon MPS, and CPU environments. You can run the automated experiment loop locally on your available hardware without requiring a dedicated GPU cluster.
Setting up a time budget per experiment requires configuring the iteration limits and time constraints in your training script execution settings. The automated loop executes each experiment within the specified time budget, evaluates the optimization metric, and retains only the improvements.
To start autonomous ML model architecture discovery, you need a configured training script path, a defined optimization metric, a time budget per experiment, and iteration limits. The loop runs continuously, modifying code and evaluating results to report the best-performing configuration.
Yes, the autonomous experiment loop modifies training code directly to iterate on model architectures and hyperparameters. It evaluates a single metric after each timed execution and keeps only the changes that result in beneficial improvements to your model.
Limitations include relying on a single optimization metric for evaluation and requiring explicit configuration of iteration limits and time budgets. The loop's effectiveness is bounded by the specified time constraints and the quality of the initial training script provided.
mehdicAssemble relevant context packages for agent spawns within token budgets.
OpenLAIRConfigure miles for large MoE RL training with FP8/INT4 and train-inference alignment.
barathanaslanOptimize PyTorch and MLX machine learning workloads on Apple Silicon.
davpatel605-beepConvert PyTorch training loops into Lightning Trainer workflows with distributed support.
