Running Chemograph Generator

Instructions for Running Chemograph Generator Note: these instructions assume the user has installed Anaconda 3 on their computer

1. Create a directory on your Desktop or in My Documents called bias factor.
2. Save the chemograph generator file (ncg_merge.py) to this directory.
3. Save the seawaveQ files to a directory of your choosing in the bias factor directory (e.g., fipronil). Make a folder in “bias factor” for each run. The code says “chem” in it right now in place of the folder name (e.g. fipronil).
4. Open the Spyder program (should be under Programs\Anaconda3 directory). The first time you open it, it may prompt you to have an administrator update something. Just close the window and Spyder should open fine.
5. Under the File menu, select open and select the ncg_merge.py code. The code for the program should open in a window. This will be where you edit areas based on your chemical and analyses.
6. Scroll down to Line 277 and in the following lines enter the name of the chemical file subdirectory you create (e.g., fipronil) a. chemograph_dir = os.path.join("fipronil") b. output_dir = os.path.join("fipronil","Output")
7. In Line 285, enter the sampling strategy you will be using. The comment in the line (text following a # sign) let’s you know the options you have.
8. In Line 286, enter the imputation technique.
9. When you are done, to run the program click the green arrow at the top or select Run, Run.
10. The status of the program run will appear in the IPython console window. This area will let you know if you run into an errors

Change sampling strategy: run stratified first, then random Linear, stair step, then log linear. Console window tells you where you are in the code/what’s being run. So if the program screws up during a run, look in the console to see why and talk to Chuck and we can probably set it up to continue the run where it left off (in a new file) End up with one output file for each chemical, each sampling strategy, and each imputation Use the mouse jiggler while doing long runs (saved under MODELS)