Seaflow is an algorithm orchestration tool developed based on directed acyclic graphs. Its main functions include algorithm workflow orchestration, automatic optimization of runtime parallel scheduling, cache optimization, and support for heterogeneity.
Project description
About seaflow
Seaflow is an algorithm orchestration tool developed based on directed acyclic graphs. Its main functions include algorithm workflow design, runtime scheduling parallel automatic optimization, cache optimization, and support for heterogeneity. Its main technologies include metaprogramming technology, Ray computing engine, and Networkx graph theory technology.
Documentation
The documentationn for user guide is at
https://seaflow-officialdoc.netlify.app
The documentation for code comments is at
https://seaflow-codedoc.netlify.app
Main Concepts
- DAGrunner: DAG executor is an execution control for Seaflow runtime, which mainly manages the actual calculation and execution of directed acyclic graphs.
- Sequencerunner: The sequence executor is an execution control for Seaflow runtime, which mainly manages the actual calculation and execution of sequence pipelines.
- Scheduler: The scheduler is an orchestration control for Seaflow runtime, providing algorithm orchestration and directed acyclic graph related algorithms.
- Datacatalog: The data directory is a storage control for Seaflow runtime, mainly providing data caching and data interaction methods.
- Hook_manager: Mount management is the plugin management for Seaflow runtime, with the main function providing dynamic plugin functionality.
- Node: Node is the basic unit of Seaflow runtime, which needs to be object-oriented encapsulated according to the objective function, supports parameter dependency, and needs to set input and output.
- Function: Function is another basic unit of Seaflow runtime that does not require object-oriented encapsulation, does not support parameter dependency, and cannot use input.
What technologies were used
How to get it
The master branch on GitHub is the most up to date code
https://github.com/redblue0216/Seaflow
Source download of release tags are available on GitHub
https://github.com/redblue0216/Seaflow/releases/tag/seaflow_distribution_wheel
Binaries and source distributions are available from PyPi
$ pip install shihua-seaflow
License
Modified BSD (3-clause)
Bug Reports
Bug reports can be submitted to the issue tracker at
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file shihua-seaflow-0.1.1.tar.gz.
File metadata
- Download URL: shihua-seaflow-0.1.1.tar.gz
- Upload date:
- Size: 18.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.9.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
00e29cf5f83ef19b3cac9fe552caf68ed8db0a04906160f10a21e6f3352bc247
|
|
| MD5 |
8eec5dfe1e0ba670a89283e0b1ab28d6
|
|
| BLAKE2b-256 |
c69bdbd09713482b809a725ff3f78fd09e630f5b6cbcf7a3749ae3698085aa43
|
File details
Details for the file shihua_seaflow-0.1.1-py3-none-any.whl.
File metadata
- Download URL: shihua_seaflow-0.1.1-py3-none-any.whl
- Upload date:
- Size: 23.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.9.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
81dc4105f0d4bea1382aac4b74feb588e5e32383780d79dd400b44b19234e69e
|
|
| MD5 |
787a517224260852d559458c6f545ccb
|
|
| BLAKE2b-256 |
c9401ca7e300999edde00cfed302405c09ce1a9d3aa3a6911606f3be8c57dce4
|