- Create an environment with Python 3.9, and install requirements (
pip -r requirements.txt
) - Execute
python start.py
and follow prompts in the terminal. - At the end of each program execution, the displayed results are saved
to a folder called
results
. The PNG contains the confusion matrix, and the CSV file with the same name contains the performance summary (accuracy et al.).
For example, if you'd like to manually filter the rows in a way that's not offered within the program itself.
Requirements:
- The CSV file must have the same column headers -- no changes to the names or order, and no adding/removing columns.
- It should use the same naming convention in the
species
column, i.e. 2 charaters for genus, 2 characters for subgenus, 2 characters for subsection, a period, and the species name.
To load your own custom CSV, specify the file path as an argument, e.g.:
python start.py data/my_new_file.csv
.
Field Museum of Natural History & Grainger Bioinformatics Center
Ryan Fuller (Postdoctoral Researcher), Beth McDonald (Machine Learning Engineer), Dr. Rick Ree (Curator & Section Head of Flowering Plants)
Please contact Ryan Fuller for research inquiries and licensing.