barn-printer-logs

There are two versions of notebooks in here:

• original one in the notebooks/ directory loading metrics from data/ directory.
• latest one, using a notebook in the ./ directory loading metrics from a Google Sheet.

New version

Using Google Colaboratory environment, the notebook Evaluate_3D_Printer_Logs.ipynb loads captured printer statistics, and generates graphs with optional downloads. It requires a Google account and permission to use it. You may not have access to the printer logs - they're in a Google Sheet I maintain in a BARN-specific account. But you can see the current version of the data captured in this notebook.

Older Version

This is a capture and quick analysis of the statistics we capture from our collection of 3D printers. There's not a lot of data available - mainly print-time and filament-used. These are cumulative values, and have (so far) recorded them intermittently.

This is also intended to be educational and demonstrative to other org members, to learn how to:

• use a Python virtual environment
• use a Jupyter notebook in the virtual environment
• use Pandas/numpy/scipy for data import, preparation, and (basic) analysis
• use Matplotlib/Seaborn for basic visualization

Local Setup

There is a requirements.txt here that captures the packages I loaded into a virtual environment, but many of them came in as transitive d ependencies. The list of packages I started with is a lot smaller and for the most part should not be version-specific. With that in mind, here's the basic instructions to setup:

• some version of Python 3 and pip should be installed.
• you should be in the directory with the repo, or where you want to install your virtual environment
• mkdir pyenv
• python3 -m venv pyenv
• source pyenv/bin/activate
• pip install --upgrade pip
• pip install wheel
• pip install pandas
• pip install scipy
• pip install matplotlib
• pip install s3fs
• pip install seaborn
• pip install scikit-learn
• pip install jupyter

--30--

Now when you're in this directory, should execute the command jupyter notebook and follow the instructions to open a browser to the appropriate location.

From there, just open the document(s) in the notebooks/ directory and run all cells.