Version for HA Core users

[PianSom]

Hi Trefor - very interested to give this a go, but I'm a HA Container (ie HA in Docker) user so can't use Add-ons. Any chance of a version which ties into the Appdeamon world, like Predbat does?

[springfall2008]

Unfortunately the appdaemon add-on itself is based on the wrong linux version so you can't use it with PyTorch.

However the add on is just a Docker container so I think you can run it in a seperate Docker you just need to pass in your HA supervisor key to the environment.

[PianSom]

Ah, pity!

And double pity - HA Container is not compatible with Supervisor. :(

[springfall2008]

I doubt its the same thing, the key I'm talking about is for the REST interface between HA and external Apps which allows them to read/write state?

[PianSom]

Oh good. I thought you meant this Supervisor interaction. But a Long Lived Access Token is very possible.

I'll have a play when I have a spare half hour, thanks.

[springfall2008]

I made it work with predbat_addon so you might be able to copy the stand-alone Dockerfile from there

[PianSom]

I'm afraid neither my GitHub nor my docker skills are up to this.

I tried docker build -t predai github.com/springfall2008/predai#main:predai
but got the error

--------------------                                                                                                                                                                 
   2 |     ARG BUILD_FROM                                                                                                                                                            
   3 |     ARG BUILD_VERSION                                                                                                                                                         
   4 | >>> FROM $BUILD_FROM                                                                                                                                                          
   5 |     
   6 |     # Set shell
--------------------
ERROR: failed to solve: base name ($BUILD_FROM) should not be blank

I think it's probably better that I wait until someone more skilled/experienced than me uploads a container to DockerHub. But thanks.

[springfall2008]

I managed it with predbat_addon so maybe I can have a go with this one soon also.

[SwiftRR]

Hi Trefor,
I have experimented with predai on my raspberry pi HAOS setup.
I have got it working and have added the relevant lines to my predbat apps.yaml.
I managed to generate an apex chart that looks very similar to the chart shown in the documentation.

I did notice that there was an extra space in the documentation code lines that I added to my apps.yaml file. It reads as below.
I removed the extra space.

  load_forecast_only: True
  load_forecast:
     - sensor.givtcp_{geserial}_load_energy_today_kwh_prediction$results 

Some questions:

1. I wasn't sure what to add for the 'subtract' line for ev use which refers to a 'sensor'. In predbat, ignoring ev charging or large usages is handled by a switch or inputs (in documentation car charging hold options). I have a zappi.
2. Should I change the historical days to 10 in the predai yaml to match what I have in HA currently?
3. To go back from using predai with predbat, do I just comment out the relevant 3 lines in my predbat apps.yaml file?

I have a heat pump with weather compensation, and I have been able to add the sensor for my weather compensation to the predai.yaml.

Rob