Import Experiment from README directly

[brosaplanella]

At the moment, you need to load the actual data to be able to extract a PyBaMM Experiment object, but you can't do it from the REDME.

The goal is to add this functionality to the readme_processor module.

[brosaplanella]

Tom Holland (@tomjholland) I get the impression that the functionality to process the README into a list of steps is already available, can you point me to where it is?

The pybamm_experiment method just extracts them from the data, but I suspect the logic to process the README is elsewhere.

[tomjholland]

Tom Holland (@tomjholland) I get the impression that the functionality to process the README into a list of steps is already available, can you point me to where it is?

The pybamm_experiment method just extracts them from the data, but I suspect the logic to process the README is elsewhere.

Hi Ferran, thanks for opening this. Yes the readme processing is done in pyprobe/readme_processor.py, which produces an experiment_dict attribute. This is then used in the RawData class (https://github.com/ImperialCollegeLondon/PyProBE/blob/main/pyprobe/rawdata.py#L262) to reconstruct a dictionary that can be passed to a PyBaMM experiment.

The latter occurs after PyProBE has filtered down your data, with the idea being to provide the pybamm experiment that corresponds to the filtered section of data. For converting the entire readme file, therefore, the readme_processor module is the place to start. The dict it builds should contain all the step instruction strings (Step Descriptions) and details of the number of repeats (Cycles).

[brosaplanella]

I have noticed some steps are CCCV (e.g. "Charge at 4mA until 4.2 V, Hold at 4.2 V until 0.04 A"), but these would be 2 PyBaMM steps. Does the conversion to PyBaMM experiment take care of it?

[tomjholland]

I have noticed some steps are CCCV (e.g. "Charge at 4mA until 4.2 V, Hold at 4.2 V until 0.04 A"), but these would be 2 PyBaMM steps. Does the conversion to PyBaMM experiment take care of it?

Yes this is to get around the fact that some cyclers do a CCCV as a single step, so the conversion separates it into two PyBaMM steps at the comma