Local setup

While Google Colab provides the easiest way to access SunSolve P90, you can also run the Jupyter notebook locally on your own computer.

This will also ensure that any output files will be saved locally.

Prerequisites

This requires installing the following programs:

• Python 3+ - Recent version of Python (3.8 or higher recommended). Download from python.org.
• Pip - Python package installer (usually included with Python). If not installed, see pip installation guide.
• Jupyter - Notebook software. Install with pip install jupyter.
• Git - For cloning the repository (alternatively, you can download the files directly). Download from git-scm.com.

Installation steps

1. Ensure Python and Jupyter are installed:

   pip install jupyter pandas numpy matplotlib plotly

2. Clone or download the repository:

   git clone https://github.com/pvlighthouse/SunSolveP90.git
   cd SunSolveP90

   Or download directly from the repository and extract the files.

3. Start Jupyter:

   jupyter notebook

4. Open the main analysis notebook (P90 Analysis.ipynb) in your Jupyter interface

5. Run the notebook: All package installation and setup is handled automatically by the notebook cells

Development environment options

Standard Jupyter notebook

Running the notebook with Jupyter directly (in-browser) requires you to:

1. Download the notebook to your local drive
2. Open a terminal in the same folder
3. Run jupyter notebook to open the Jupyter interface
4. Open the “P90 Analysis” notebook from this interface
5. Select a valid kernel (usually ‘venv’ or ‘conda’)

After this point the experience is mostly the same as for Colab. Just be aware the play button is under the top menu, instead of on the left of the cells.

Visual Studio Code

If you prefer VS Code for Jupyter notebook development:

1. Install the Python extension and Jupyter extension
2. Open the cloned repository folder in VS Code
3. Select the Python interpreter with your installed packages
4. Open P90 Analysis.ipynb and run cells interactively

VS Code users will see a very similar interface to Google Colab, with play buttons also on the left. Simply save the notebook to your local drive and then open it in VS Code.

The main difference is that when logging in, you will see the prompt in your search bar at the top of your screen.

JupyterLab

For a more advanced interface:

1. Install JupyterLab: pip install jupyterlab
2. Start with: jupyter lab
3. Navigate to the P90 notebook files in the interface

Troubleshooting

If you run the Jupyter notebook locally without Google Colab, you will need to install all the required modules manually.

If you encounter a “ModuleNotFoundError”, run a cell with the code %pip install ___ replacing the ”___” with the name of the missing module.

Authentication setup

During your first analysis, you’ll be prompted to authenticate:

1. Credential Prompt: The system will request your PV Lighthouse login credentials
2. VS Code Users: Credentials will be requested in the search bar
3. Other Environments: Follow the interactive prompts in your notebook

Your credentials are used only for API authentication and are not stored permanently.

Next steps

Once your environment is set up, proceed to the Jupyter notebook walkthrough to run your first uncertainty analysis.