quick_start.rst•2.08 kB
Quick Start Guide
=================
This guide will get you up and running with RMCP in minutes.
Starting the Server
-------------------
1. **Install RMCP**:
.. code-block:: bash
pip install rmcp
2. **Check R packages**:
.. code-block:: bash
rmcp check-r-packages
3. **Start the MCP server**:
.. code-block:: bash
rmcp start
Basic Usage Examples
--------------------
Linear Regression
~~~~~~~~~~~~~~~~~
.. code-block:: json
{
"tool": "linear_model",
"data": {
"x": [1, 2, 3, 4, 5],
"y": [2, 4, 6, 8, 10]
},
"formula": "y ~ x"
}
Correlation Analysis
~~~~~~~~~~~~~~~~~~~
.. code-block:: json
{
"tool": "correlation_analysis",
"data": {
"var1": [1, 2, 3, 4, 5],
"var2": [2, 4, 6, 8, 10],
"var3": [1, 3, 5, 7, 9]
}
}
Visualization with Inline Display
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. code-block:: json
{
"tool": "scatter_plot",
"data": {
"x": [1, 2, 3, 4, 5],
"y": [2, 4, 6, 8, 10]
},
"x_var": "x",
"y_var": "y",
"return_image": true
}
Working with Real Data
----------------------
Load Example Datasets
~~~~~~~~~~~~~~~~~~~~~
.. code-block:: json
{
"tool": "load_example",
"dataset_name": "sales",
"size": "small"
}
Import Your Data
~~~~~~~~~~~~~~~
.. code-block:: json
{
"tool": "read_csv",
"file_path": "/path/to/your/data.csv"
}
Natural Language Formula Building
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. code-block:: json
{
"tool": "build_formula",
"description": "predict sales from marketing spend"
}
Error Recovery
~~~~~~~~~~~~~~
.. code-block:: json
{
"tool": "suggest_fix",
"error_message": "object 'sales' not found",
"tool_name": "linear_model"
}
Next Steps
----------
* :doc:`../troubleshooting` - Troubleshooting common issues
* :doc:`../api/modules` - Complete API reference