invivoinfer is a python package containing the code described in Extraction of Synaptic Input Properties in-vivo (2016) P.Puggioni, M. Jelitai, I. Duguid, M. van Rossum
** You should feel that the code is undertested and certainly sub-optimal **
- Clone this repo and run (it will install all the dependencies too)
git clone https://github.com/ppuggioni/invivoinfer.git cd invivoinfer pip install -r requirements.txt python setup.py develop
- if you want, run the tests in the test folder, to make sure all is installed correctly
- open jupyter notebook
jupyter notebook
and open the notebook notebooks/Example_analysis.ipynb
if you run the notebook, you should get all the plots and understand how to use the package. Note that running the notebook as it is might be long (~15/20 minutes?). For testing purposes, you should run with config_testing.json when initialising the class (at some point in the notebook I wrote a warning).
This is the main file where you control the options of the inference. Probably the most important one at the beginning is the one to baseline the trace:
"baseline_corr": {
"average": 95,
"uncertainty": 5,
"ToUse": true}
Where average is the baseline and uncertainty is, as you expect, the uncertainty of your baseline estimation.