MCP Machine Learning Agent
Related Servers
Alternatives to MCP Machine Learning Agent
No user-submitted related servers found.
Related Servers
- AlicenseAqualityBmaintenanceEnables training, evaluating, and comparing Logistic Regression and KNN models on the Wisconsin breast cancer dataset, then predicting malignant or benign diagnoses from patient feature inputs and generating SHAP explanations for the selected model. Every tool call is logged to a Parquet audit trail that can be exported for review.9GPL 3.0
- AlicenseAqualityCmaintenanceEnables users to train and compare Logistic Regression and KNN models on the Wisconsin breast cancer dataset, run predictions on patient feature inputs, compute confusion-matrix metrics and select the best model by malignant recall, generate global and local SHAP explanations, and export Parquet audit logs of every tool call. Also provides sample test-set inputs so agents can invoke prediction tools with valid 30-feature payloads.9GPL 3.0
- AlicenseAqualityBmaintenanceExposes MCP tools that train, evaluate, compare and explain Logistic Regression and KNN models on the Wisconsin breast cancer dataset, then serve malignant/benign predictions with SHAP explanations and a Parquet audit log of every call. It also provides ready-to-use sample patient inputs and model-selection tooling so agents can run a full ML workflow through natural language.9GPL 3.0
- AlicenseAqualityBmaintenanceEnables training and evaluating Logistic Regression and KNN models on breast cancer data, making predictions, selecting the best model, generating SHAP explanations, and exporting Parquet audit logs through MCP tools.9GPL 3.0
- AlicenseAqualityCmaintenanceEnables training and comparing Logistic Regression and KNN models, making breast-cancer predictions, evaluating confusion-matrix metrics, selecting the best model, generating SHAP explanations, and exporting a Parquet audit log through MCP tools.9GPL 3.0
- AlicenseAqualityCmaintenanceEnables training and comparing Logistic Regression and KNN breast-cancer classifiers, getting sample inputs and predictions, evaluating confusion-matrix metrics, selecting the best model, generating SHAP explanations, and exporting a Parquet audit log through MCP tools.9GPL 3.0
TDQS
Scored across 8 tools
Each tool has a clearly distinct role in the ML workflow: training, sample retrieval, per-model prediction, audit export, metrics, model selection, and SHAP explanation. The only mild overlap is between confusion_matrix_metrics and select_best_model, but their descriptions make the boundary clear.
Most tools follow a verb_noun snake_case pattern (train_models, get_sample_input, predict_logistic_regression, export_parquet_audit, select_best_model). Two names deviate: confusion_matrix_metrics is noun-based, and shap_explain_selected has an unusual noun_verb order.
Eight tools is well-scoped for a focused ML agent covering training, prediction, evaluation, selection, explanation, and audit export. Each tool earns its place without obvious redundancy.
The surface covers the core ML lifecycle: train, sample input, predict with both models, evaluate, select best, explain, and export audit. A minor gap is the lack of a unified predict_selected_model tool, requiring the agent to manually choose the corresponding per-model predictor after selection.