TonyBogomoloff

import numpy as np, pandas as pd

def markBlocks(data, condition, blockName):
    df = data.copy()

    ### count the length of periods where the condition is satisfied
    df['flag'] = np.where(condition, 1, 0)
    ## give a consequetive number to each step, where the the condition is satisfied
    df['counter_' + blockName] = df.groupby((df['flag'] != df['flag'].shift(1)).cumsum()).cumcount()+1
    ## drop the counter to zero when the condition is not satisfied

TonyBogomoloff

qwerty

TonyBogomoloff

wr_device = device.Device(serial="WR2870546688") ## the one from Gautier
command_set = 1     # b'\x00\xe6\x00' # https://my.tado.com/support/index?home.id=21004

subset = metadata[metadata.command_set_code == command_set]
print('Number of commands in the subset:', len(subset))
cmd = subset.ir_data.values[1]
cmd = 'AF12D1A832B2121255540E65556F1A63E7586A07A96FB26B000053FF5500818B210001646110115580116164640065726570716F01111110001111010321000100001011011111001000000000011001011100001000000000001111111011111110111100001010000000000000101010054210001000010110111110010000001111101100001111110001111010321000100001011011111001000000000010110000111111000111101054'

cmd = metadata.ir_data.values[0]
org = 0 if metadata[metadata.ir_data == cmd].command_set_code.values[0] < 10000 else 1

TonyBogomoloff

### Keybase proof

I hereby claim:

  * I am tonybogomoloff on github.
  * I am tonybogomoloff (https://keybase.io/tonybogomoloff) on keybase.
  * I have a public key ASA3STRH5S3DoF5p9Y0bNoR0YKP-bTmq47gj7kKwGfdVbwo

To claim this, I am signing this object: