ML expert system: fundamentals, training pipelines, deployment, and MLOps
Ready-to-adapt code for sklearn pipelines, PyTorch training loops, FastAPI model serving, and MLflow experiment tracking — plus best practices and an anti-patterns list (no data leakage, no untracked models)
- What
- Ready-to-adapt code for sklearn pipelines, PyTorch training loops, FastAPI model serving, and MLflow experiment tracking — plus best practices and an anti-patterns list (no data leakage, no untracked models)
- Cost
- Free
- Needs
- a Python ML project to apply it to; MLflow server and serving infrastructure only if you want the tracking/deployment parts — the skill is guidance plus starting-point code, not a package
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor — @personamanagmentlayer's ml-expert skill: an expert-level ML system covering the whole lifecycle — fundamentals (supervised/unsupervised/reinforcement, feature engineering, hyperparameter tuning), deep learning (CNNs/RNNs/Transformers, transfer learning, attention, GANs, autoencoders), and MLOps (versioning, experiment tracking, CI/CD, monitoring, retraining, A/B testing) — with four ready-to-adapt code systems: a sklearn `MLPipeline` class (stratified split, scaling, random forest with CV, evaluation, joblib persistence), a PyTorch `NeuralNetwork` + `Trainer` (train/eval loops, device handling), a FastAPI `ModelServer` (prediction endpoint with pydantic models, health check), and an `MLflowExperiment` wrapper (run logging, model registry, production promotion). Includes best-practice lists per phase and an anti-patterns list: no training on test data (data leakage), no missing validation set, no ignored class imbalance, no unscaled features, no model versioning gaps, no unmonitored production. Honest caveats: the skill declares `allowed-tools` (Read/Write/Edit, python-only Bash) — in agent runtimes that honor them; code is starting-point quality, adapt model choices and hyperparameters to your data; MLflow/deployment bits need their own infrastructure. Apache-2.0 licensed. Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version:
Install
Prerequisites: a Python ML project to apply it to; MLflow server and serving infrastructure only if you want the tracking/deployment parts — the skill is guidance plus starting-point code, not a package Install "ML expert system: fundamentals, training pipelines, deployment, and MLOps" for me. It gives my agent @personamanagmentlayer's ml-expert system: ML fundamentals and anti-patterns, four ready-to-adapt code systems (sklearn MLPipeline, PyTorch NeuralNetwork+Trainer, FastAPI ModelServer, MLflowExperiment wrapper), best practices per phase, and MLOps discipline (versioning, monitoring, retraining). Apache-2.0 licensed. Repository: https://github.com/personamanagmentlayer/pcl/blob/main/stdlib/ai/ml-expert/SKILL.md 1. Fetch the SKILL.md file (and any helper files) from the repository path into a temporary folder and summarize what it does in one or two sentences. 2. Safety check: review the SKILL.md and scripts for anything suspicious (unexpected network calls, shell commands, credential harvesting). This repo should contain zero secrets in code, credentials only via the secure vault, allowed hosts declared in the SKILL.md. Verify that holds here; STOP on any red flag and tell me. 3. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "ml-expert". 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (e.g. point the agent at my ML task and dataset; the skill declares allowed-tools (Read/Write/Edit, python-only Bash) — check my agent runtime honors them; adapt the starter code's model choices and hyperparameters to my data; provision MLflow/serving infra only for the tracking/deployment parts). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.