Getting Started¶
Requirements¶
Python 3.7.+
Packages:
numpy
scipy
bokeh
xlrd
pandas
numpyandscipyare basic requirements.bokehandxlrdare needed for plotting and/or GUI.pandasis needed for reading data from Excel in one of the examples.
It is generally recommended to use python virtual environment or conda virtual environment.
Setup¶
Just the library
Install via pip package manager:
pip install bio_rtd
Source code with the examples
Clone git repo:
git clone https://github.com/open-biotech/bio-rtd.git
Go to repo folder:
cd bio-rtd
Install requirements:
pip install -r requirements.txt
Running first example¶
Run an example:
python example/models/single_pcc.py
If you see:
ModuleNotFoundError: No module named 'bio_rtd'
then you need to add repo to PYTHONPATH (either in your IDE or in terminal):
PyCharm
Check "Add source root to PYTHONPATH" under
"Run/Debug configuration".
Terminal
export PYTHONPATH=${PYTHONPATH}:`pwd`
Running example with bokeh serve¶
Examples that end with _gui.py are
python scripts for creating an interactive web application.
In background a Bokeh Server instance is created. The server connects UI elements with python script (see more at Building Bokeh Applications).
Terminal
bokeh serve --show example/models/integrated_mab_gui.py
bokeh can also run as a python script (good for debugging):
python `which bokeh` serve --show example/models/integrated_mab_gui.py
PyCharm
To run python bokeh serve with PyCharm, set the following Run Configuration:
Configuration: python
Script path: /path_to/bokeh
Parameters: serve --show /example/models/integrated_mab_gui.py
where /path_to/bokeh can be obtained by running which bokeh command in terminal in PyCharm.
Also make sure the repo is added to the PYTHONPATH as described in Running first example.
Flag --show is optional and runs the newly created instance in a new tab in a web browser.
Only one server instance can be run (on the same port). If you try to run another bokeh serve command while one
is running, you will see the following exception:
Cannot start Bokeh server, port 5006 is already in use
In that case find and close the existing running bokeh serve process.