Timeseries to DataFrame

In Kelvin SmartApps™ you can retrieve timeseries data from the Kelvin Cloud for use in your programs or for training machine learning models.

Data Scientists commonly will use the Pandas DataFrames format for managing large amounts of data due to its flexibility and data management capabilities.

Kelvin API Client (Python) has a build in utility to easily convert any timeseries data into a DataFrame.

from kelvin.api.client import Client
from kelvin.api.client.dataframe_conversion import timeseries_to_dataframe

# Login
client = Client(config={"url": "https://<url.kelvin.ai>", "username": "<your_username>"})
client.login(password="<your_password>")

response = client.timeseries.get_timeseries_range(data={
  "agg": "mean",
  "end_time": "2024-04-18T16:35:44.703732Z",
  "fill": "none",
  "group_by_selector": True,
  "order": "ASC",
  "selectors": [
    {
      "fields": [
        "value"
      ],
      "resource": "krn:ad:demo_asset_02/demo-data-stream-01"
    }
  ],
  "start_time": "2024-04-18T12:35:44.703732Z",
  "time_bucket": "1h"
})

df = timeseries_to_dataframe(response)