Today we are launching our brand new data science portal to teach you how you can learn from your own smart home data. Learn what data Home Assistant stores locally and learn to explore your data. In 15 minutes, you will set up a local data science environment and are running reports on your own data.
We have worked hard on a smooth experience. Single click to set up a local data science environment powered by Jupyter. Pre-installed with our Getting Started notebook for Home Assistant, automatically connects to your database and generates reports with a single click.
All of this is powered by the new JupyterLab hass.io add-on by @frenck and the new HASS Data Detective Python package by @robmarkcole.
Today is only the start. We will continue to develop the data tools for Home Assistant. It’s our goal to make Home Assistant the easiest platform to use for data scientists interested in home automation.
its a data science portal for HASS that is only released as an addon.
and i think its a very bad sign if there are things created for home assistant that are not usable for not hassio users.
hassio is build for the less experienced users that are less able to program.
hass is grown because of the more experienced users that are able to program.
and a lot of those users dont use hassio, but they are the most important part from this community.
as soon as those people feel excluded, or forced into an environment like hassio which restricts them, they will look elsewhere and stop supporting hass.
and without the community hass isnt halve as great as it is.
i can expect and accept when third parties build and offer new functions to hassio, that are not available to people without hassio, but i really hope that the hass devs will stay far away from that, because it will destroy hass.
I just think it’s uncool to announce a new feature that everyone could use and limit it to just a HASS.io add-on. I’ve never heard of this and a few key pointers to get it setup would go a long way.
You can build your own juypter server or run a docker image of one. thats what i have done… the problem i have at moment is the notebooks as i cant seem to run them.
i dont even expect a complete guide how to install juypter, but i think the guide from HA should say which parts need to be done yourself (preferably with some links) and then which parts of the guide are usefull for none hassio users.
for many components that is simular ( a link to some webpage or even a guide how to setup webservices and even parts that need to be installed sometimes)
edit: by the way noone should think that i ask shout out for this for myself. i dont use the database from HA because i think it saves to much and is setup completely wrong thats why it grows big very fast.
but i am really concerned about the way this is going, because i love HASS and i would love to see it get really great, and a fear the worst when i see stuff like this.
This is only day 1 of this new feature, it already mentions its only the start. Lets give them a chance to start getting it out there before jumping up and down and saying were being left out.
If that does happen, then yes we can start shouting
maybe you are right, but i think its a very bad sign in general.
it should be the other way around, create options and docs first and then, create an addon for it. (or release both at the same time)
I dont think its a bad sign, not yet, if yes it stays just a hassio platform and other items do start following that excludes some of us, then yes we can feel upset.
I think there is some confusion here. You can install the HASS-data-detective on any platform as its just a python package, i’m actually running it on my Mac. The Hassio addon is for Jupyterlab, but likewise you can install and use Jupyterlab on any platform also. Please refer to the Jupyterlab docs for instructions for your platform. You can create the env the addon uses by installing these requirements.
Cheers
When running the cells within the ‘GETTING STARTED’ workbook I recieve an error: ValueError: Secrets file /config/includes/secrets.yaml not found.
My ‘secrets.yaml’ file is directly within ‘/config/’ directory, is there a parameter I can adjust?
I agree with all the upset people in the thread, especially the comment that it should be the other way round. Build something for Homeassistant, then make an addon so it can be easily integrated into hassio.
Some of the core promises that were made about this project not becoming reliant on addons and always being able to be dealt with in code seem to be being eroded, and it is a very sad time IMO.
It might be a bit discouraging at first to read that it is powered by a hass.io add-on, but it is actually not too complicated to set it up on any other platform supported by Jupyter. The hass.io add-on simply comes pre-configured and is ready to use.