Skip to content

Commit

Permalink
Merge pull request #29 from finalTestin/dev-0.2.1
Browse files Browse the repository at this point in the history
update README
  • Loading branch information
alexzyWu authored Jul 29, 2021
2 parents 99e3783 + 8786f38 commit ee61f21
Show file tree
Hide file tree
Showing 5 changed files with 35 additions and 10 deletions.
4 changes: 4 additions & 0 deletions cc/README.md
Original file line number Diff line number Diff line change
@@ -1,2 +1,6 @@
# MLSS-ControlCenter-Go



CC is the platform control subsystem of the Prophecis system, which mainly includes basic platform management, user information management, project group information management, namespace management, machine tag management, and data management. Users need to apply for resources first, and then carry out model development and training.

19 changes: 10 additions & 9 deletions di/README.md
Original file line number Diff line number Diff line change
@@ -1,17 +1,18 @@
### Prophecis-DI module is built based on the [FfDL](https://github.com/IBM/FfDL).
# Prophecis-DI

**The main modifications are as follows**:

[1] Integrate Kubeflow Arena,Provide distributed tensorflow task ability.

[2] Modify the creation mode of single machine modeling task:remove helper and job jobmonitor in task, and change deploy pod to deploy job.
DI is the model training execution subsystem of the Prophecis system. It supports users to choose a stand-alone or distributed method to train the model. Users can access the CPU and GPU resources required for training through the UI. The DI subsystem will allocate according to the information configured by the user. Corresponding resources.

[3] The log collection service is changed to daemonset, and the collection tool is changed to fluent bit.

[4] The task status update mode is changed to an independent service job monitor.

[5] Add user GUID control in container data directory.
#### Module

[6] Enhance CLI, added parameter replacement of yaml template, and the train command was modified to websocket connect, providing log and state.
- CLI: CLI is a command line tool, which can perform basic command operations on Prophecis-DI.
- Jonmonitor: Jonmonitor is responsible for monitoring the status of task execution.
- LCM: LCM is responsible for various methods of task life cycle management, including start, pause, and terminate.

- Restapi: Restapi is mainly used to process external service requests and service authentication, and supports load balancing.
- Storage: Storage is responsible for the operation of storage module, such as Minio, ES, Mongo, etc.
- Trainer: Trainer is responsible for managing model training tasks.

[7] The code file storage server is changed to Minio.
6 changes: 6 additions & 0 deletions linkismagic/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
# LinkisMagic



LinkisMagic is a set of defined Magic functions, which are used to calculate the interaction with the computing middleware Linkis, and Linkis submits the calculation engine. The Magic function sends the Code to Linkis via Http request, and Linkis returns the result and displays it in Jupyter Lab.

6 changes: 6 additions & 0 deletions mllabis/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
# MLLabis



MLLabis is the personal development environment subsystem of the Prophecis system. It allows users to create Notebooks to edit codes and develop models online. It also supports the integrated development environment of Hadoop and Spark to facilitate user development.

10 changes: 9 additions & 1 deletion ui/README.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,14 @@
# Prophecis-UI



UI is the front-end part of the website of other subsystems of prophecis, including front-end pages and page jump controllers, and is mainly responsible for the view control and data display of users visiting web pages.



# new-project

## Project setup
### Project setup
```
npm install
```
Expand Down

0 comments on commit ee61f21

Please sign in to comment.