New structure of directories

[apascualgarcia]

@jordan-klein (@Jennifer-Villers FYI)

I performed the following changes in directories and files:

• I created a set of new directories at real_models to make easier to classify all the simulations and give them more informative names (for running the models massively it will be less prone to errors). In principle the old model_1_$ could be removed, but double check first.

• Each directory is a combination of a type of population structure and a type of contact matrix, which are the inputs for my script. Therefore all simulations with a given population structure and contact matrix will be stored in the same directory, and I will create a new directory for each simulation.

• Each directory is labelled with the parameters used, and each file contains that label.

• Since every time there is a change in the population structures or in the contact matrices, we should update them in all these directories, I am going to try a different strategy. All population structure files and contact matrices will be stored in data/estimation_parameters and I will create symbolic links in each directory where the simulations are run pointing to them. Let's see how this works with git, all the symbolic links will be relative. Could you please check if you see the symbolic links and that they are pointing to the correct files for the new directories called "shield_$params". If you can then do the same with the "null_model" dirs would be great. Note that for the shielded populations with different % of population shielded, the contact matrices should be recomputed.

• I assume that after your changes in the population structure I had to run again the computation of the contact matrices. The new matrices are also located at data/estimation_parameters (note that I call them now contact matrices, I was messing up with the contacts_structure ones). The files contacts_structure were also moved to that directory and renamed.

[jordan-klein]

@apascualgarcia The symbolic links for the directories you created all look like they point to the correct files.

[apascualgarcia]

@jordan-klein if your scripts within model_1_1 and model_1_2 are safe please git rm these directories because are no longer needed.

Please note that I also moved the files in data/estimation_parameters/class_structured_data/shielding_scenarios_population_structure/ to data/estimation_parameters/class_structured_data/ to have them all in the same location, I created the /classes_contacts_shieldage$ files that were missing and I am going to compute contact matrices for them. I keep the (now empty) directory in case you have any script pointing there, but you can git remove it otherwise.

Please close this issue when you are done.