Experiment Notebook (enb) user manual

|enb| logo

The Experiment Notebook (from now on, enb) library is designed to help you obtain and report computer-based experimental data. Focus on what is new in your current experiment, and let enb deal with the repetitive stuff.

Some of the main features of enb are:

  • Flexible: With a few lines of code, you can easily create custom experiments and/or analyze the results you want. enb provides an easy Python interface and out-of-the-box templates and examples and this user manual to get you started in no time. By default, only standard packages need to be installed.

  • Disseminable and reproducible: enb offers simple and complex tools to create figures and tables to report your data. You can also distribute your project folder and let others verify and re-run your experiment.

  • Persistent: With enb, all intermediate results are periodically stored and kept after an experiement is completed. This way, you don’t need to run all computations over and over again. This applies to others running your experiments on other computers.

  • Extensible: you can always add new samples, define new columns and produce new figures. You will only need to obtain results for the newly defined data columns or samples.

  • Parallel in nature: you can use all the CPUs / GPUs in your local machine, and even distribute the load across a cluster of computers (clusters only supported on Linux and MacOS).

Please visit the enb github page for full access to the code.

Note

enb is based on well-known python libraries such as pandas and matplotlib. No prior knowledge of these libraries is required, although basic understanding of pandas can be most useful.

Note

Clustering support is provided only on Linux and MacOS systems. The ray library as well as the ssh and sshfs tools are employed for this purpose. If ray is not available, e.g., on Windows, the pathos library is employed for parallelization, and no clustering support is provided. See full details for multi-computer parallelization in Configuring enb in a cluster of computers.

The following help pages will tour you through the most important features of enb, and show minimal examples that you can use as starting point for your projects. You can also take a look at the automatically generated API.

Documentation contents: