I have weather data in a python dict that I'm trying to convert to a pandas df. (From there I'll load it into SQLServer but I've got that part working) my_dict = { 'data.outdoor.temperature': {'unit': '℃', 'list': {'datetime.datetime(2025, 4, 23, 10, 0): 22.3', 'datetime.datetime(2025, 4, 23, 14, 0): 21.3', 'datetime.datetime(2025, 4, 23, 18, 0): 18.2', }}, 'data.indoor.temperature': {'unit': '℃', 'list': {'datetime.datetime(2025, 4, 23, 10, 0): 23.2', 'datetime.datetime(2025, 4, 23, 14, 0): 23.5', 'datetime.datetime(2025, 4, 23, 18, 0): 22.9'}}} The output I'm after is: outdoor.temperature | 2025_04_20 | 14 | 28.1 | ℃ outdoor.temperature | 2025_04_20 | 18 | 23.8 | ℃ .....etc indoor.temperature | 2025_04_20 | 14 | 23.5 | ℃ indoor.temperature | 2025_04_20 | 18 | 23.8 | ℃ .....etc The list of datetime & temperatures could be 100s of records I've tried from_dict, concat / unstack, normalize, flatten and combinations of the lot but I'm beaten Continue reading...
