netdata-insights-alt 0.1.21

Python library for Netdata 'Insights' features.

Netdata Insights

Netdata Insights.

Install

pip install netdata_insights

Quickstart

import random 
import pandas as pd
import numpy as np

from netdata_insights.correlations.core import run_model_dict
from netdata_insights.data.utils import get_data_dict, get_chart_list

# pick some charts
#charts = random.sample(get_chart_list(host='london.my-netdata.io'),2)
charts = ['system.cpu']

# get some data
dict_baseline = get_data_dict(after=-120, before=0, host='london.my-netdata.io', charts=charts)
dict_highlight = get_data_dict(after=-560, before=-120, host='london.my-netdata.io', charts=charts)


# run correlations
results = run_model_dict('ks', dict_baseline, dict_highlight)
results

{'data': {'system.cpu': {'softirq': {'score': 0.0661},
   'iowait': {'score': 0.0071},
   'guest': {'score': 0.0},
   'nice': {'score': 0.0},
   'steal': {'score': 0.0405},
   'system': {'score': 0.1518},
   'irq': {'score': 0.0},
   'guest_nice': {'score': 0.0},
   'user': {'score': 0.2244}}},
 'summary': {'model_level': 'dim',
  'n_charts': 1,
  'n_dims': 9,
  'n_bad_data': 0,
  'fit_success': 9,
  'fit_default': 0,
  'fit_fail': 0,
  'success_rate': 1.0,
  'bad_data_rate': 0.0,
  'n_lags': 0,
  'fit_manual': 0}}