Skip to main content

Function manager

Overview​

Functions are the building blocks of the Function manager feature. It comprises of user-defined python notebooks that perform unique tasks. The user can access the functions under the Lab icon on the Altergo platform. Following are the predefined guidelines for the user while working with the functions

  • Upload function files as python notebooks (.ipynb-format)
  • Entail the first cell of the notebook with all mandatory inputs for the function to run
  • If you want to give dynamic parameters, make sure to add parameters in a cell tagged parameters

function_overview.png

Create​

To create a new function, follow these steps

  1. Navigate to Lab > Functions
  2. Click on Create function

create_function.jpeg

  1. Add a function Name
  2. Select the Type of file as Notebook
  3. Upload notebook file (ipynb-format)
  4. Click on Create function to save

The user is free to choose between Link to file and FIle upload as the source of function to be integrated. By default, it is selected as File upload

![function_source_fileUpload.jpeg](/migration/images/function_source_fileUpload (3).jpeg)

With this option you are able to link a notebook file hosted on a public repository (github, gitlab, bitbucket, etc...) To import notebook from a public repository, follow these steps

  1. Select Source as Link to file
  2. Add link to the file in Path
  3. Click on Create function to save

![function_source_link.jpeg](/migration/images/function_source_link (1).jpeg)

Limitations​

Depending on the configuration of the server where the notebook is executed, there are certain limitations in terms of the memory that a notebook can consume.

Here are the limitations for a server with the following configurations : CPU - 1 RAM - 2GB

caution

Maximum memory consumption - 500 MB Approx total no of datapoints - 500000 (0.5 million) Approx time range of data* - 2 months

'*' - The calculation is based on the assumption that the data is processed for a single sensor of a single asset having roughly a frequency of 1 datapoint every 10 seconds. The limits will vary based on the no of assets, no of sensors and the frequency of the data on Altergo.