http://allofphysicsgraph.github.io/proofofconcept/
Physics Derivation Graph
- Claim: a finite static directed graph exists which describes all of mathematical physics.
- Claim: the graph representation is machine parsable
- Claim: this graph can be checked by a computer algebra system
Creative Commons Attribution 4.0 International License
Python packages:
- yaml
Non-Python applications
- Latex
- Graphviz (neato)
main commands:
cd v3_CSV/
python bin/create_picture_per_derivation_from_connectionsDB.py
tools
python bin/list_derivations.sh
python bin/list_inference_rules.py
statistical analysis
python bin/popularity_of_inference_rules.py
python bin/popularity_of_expressions.py
python bin/popularity_of_symbols.py
validate connections
python bin/check_connections_using_CAS.py
Command:
python bin/build_connections_graph.py
output: two pictures, each showing the graph for the selected sequence
- "out_no_labels.png"
- "out_with_labels.png"
These were generated using GraphViz and the output file "connection_result.gv" and the command
neato -Tpng connections_result.gv > out_with_labels.png
The other two output files are the same graph, "connections_result.py" for use with Networkx "connections_result.graphml" for input to graph plotting programs like yED
Development and testing performed on Mac OS 10.11.2
Currently, content is entered into CSV files:
- connections_database.csv: relation between inference rules and expressions ** one inference rule per step (for Latex generation)
- expressions_database.csv : library of unique expressions used by connections
- inference_rules_database.csv : library of unique inference rules used by connections
- comments_database.xml : set of comments which may reference multiple expressions or steps
doc/physics_graph_notes.log: my commentary and observations on the project as it progresses