This is the same release as First releas: paper version, just that it is supposed to be archived also on Zenodo.

Essential files

For trying out the software, I recommend downloading (instead of all 7GB results) only a few following files in the top folder (where exact_ed.py is located):

• exact_ed.py
• diophantine_solver.py
• doones.py
• linear_database_newbl.csv
• cores_test.csv
• sindy_oeis.py
• blacklist.py (redundant, feel free to remove it from the code)

The list above is not checked so please make sure there are no import errors.

Get essential files via git

Optionally use git to automatically download the essential files to try out the method: git clone -n --depth=1 --filter=tree:0 https://github.com/B0Gec/Diofantos cd Diofantos git restore --source HEAD exact_ed.py diophantine_solver.py doones.py cores_test.csv sindy_oeis.py blacklist.py In the end download linear_database_newbl.csv manually since it is stored as git lfs (large files).

And ignore files under the GitHub "Assets" section.

Simple example

of execution in terminal:

python doones.py --task_id 14 --exper_id output_dir

will produce the output file results/output_file/00014_A000045.txt with similar content:

While total time consumed by now, scale:15/164, seq_id:A000045, order:10 took:
 0.6 seconds, i.e. 0.01 minutes or 0.0 hours.
CORELIST True, SINDy False, GROUND_TRUTH False, SINDy_default True
Library: n, max_order 10, threshold: 0.1

by degree: 1 and order: 2. 
A000045: 
a(n) = a(n - 2) + a(n - 1)
truth: 


  -  checked against website ground truth.     
True  -  "manual" check if equation is correct.