Setup
Install knowledge-graph using these commands:
git clone https://github.com/hyroai/knowledge-graph.git
pip install -e ./knowledge-graph
pip install cloud-utils@https://github.com/hyroai/cloud-utils/tarball/master
Publish to PyPI
After merging your new branch to the main branch, You need to publish a new version to PyPI. On the repo's main view go to the "Code" tab on top and then press on releases. Click "Draft a new release" and create you new release with semantic version that follows the rules here: https://semver.org/. Publishing your new release will create a new version on PyPI using the latest commit on main.
When to use querying.py vs querying_raw.py
Both modules provide querying abilities, but querying.py relies on a global store of kgs, storred by their hash value. The code retrieves the kg instance from this map ad hoc. This was built so we can have a serializable representation of a node in the kg, Node, which has graph_id and node_id, and not have to serialize the entire kg any time we serialize an object containing them (e.g. if we're serializing or hashing a NounPhrase for some reason).
The global store is an impure component that can make some code have different behaviours depending on what happened before (whether or not the kg was loaded).
Consequently, in contexts where you have a Element and KnowledgeGraph, prefer using querying_raw.py, giving the graph instance explicitly. This is mainly when creating graphs or enriching them.
In contexts where all you have is Node, use querying.py.
If you implement new querying functions, you can implement them in querying_raw.py, and lift them to querying.py using storage.run_on_kg_and_node.
Do not use the internal modules inside the knowledge_graph directory
Instead, rely on the imports in knowledge_graph/__init__.py, possibly add what you need there (in rare cases this is required).
Relations naming conventions
Relation are the middle part of the triplet. Although they can safely be named anything, we use a naming convention that serves as mnemonic device. For example the relation person/gender implies the left hand side is an entity representing a person, whereas the right hand side is a gender, so we can expect a triplet like Alice,person/gender,female.
Release files for knowledge-graph 0.0.35
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| knowledge_graph-0.0.35.tar.gz | 17.0 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| knowledge_graph-0.0.35-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 36.8 kB
Release files / knowledge_graph-0.0.35.tar.gz
| Download URL | knowledge_graph-0.0.35.tar.gz |
|---|---|
| Size | 17.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
936def5c880a668e0ae38aea8f0b1bfcac237fd11a59e9fcf0d509e68cf169d9
|
|
BLAKE2b-256 checksum How to use checksums |
51ff98d56eaf93dce59ce028eea4647a091863bf482773e667702d1ce79a3b7f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.14.6
|
Release files / knowledge_graph-0.0.35-py3-none-any.whl
| Download URL | knowledge_graph-0.0.35-py3-none-any.whl |
|---|---|
| Size | 19.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
c7e35553085023c09afcc94c7d42a3b3bae0969db6b1576df61d6fe630f799d1
|
|
BLAKE2b-256 checksum How to use checksums |
4066289d6f7785678f4b567eee8cde0e3f7ab4696144fb448c596a918aa377d7
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.14.6
|
