How did you compare, e.g I find it quite hard to compare 50% chance of rain…either way this is correct 
EDIT (now 19.5) it is raining slightly and heavily clouded, the ‘regular’ forecast shows that but below it is different. Yesterday I had reverse info…aargh.
The rain probabilities of the forecast are just estimated by me by analysing the text forecast. This is also where the changes for the tweaked version V1.1 happened. For now the most accurate should be the text forecast. (the first sunny is also just the state of the pressure and doesnt represent the current condition!
The forecast is also pretty depandant on wind direction do have set this up? This could explain the settled fine. This sould mean like rain is just over. For me this is only there when there is no wind!
For me these short term forecasts also nearly never match with reality.
For me today the V1.1 with wind data matches exactly reality!
It is cloudy and getting better. The forecast says:
If you want to tweak the icons and rain percentage you can look here in the code:
As you can see the chance of rain beening 50% is only on one specific condition with the latest version!
I also now use my own ecowitt integration to get even better data locally you can look here if you are interested:
OK, will try v1.1 later today.
The wind direction is from meteofrance.
With curren 0.2, the card changed a bit unexpected … times moved about 2.5h backwards in time
Note that I am only providing the info so you know, not to ask for improvements (only if you know how)
ok thats strange. I actually don’t know how this can even happen. Maybe a reboot of HA fixes it 
Still checking and for now ‘your’ sensor is more wrong then right, sadly…not saying the regular ones are better.
A few times forecast_short_term was not good, I have 1026 now and this leads to “sunny” where it is raining quite a lot. I think I will remove this one as it also has no added value, I can SEE current condition 
I will continue comparing
Found an error in row 85: this needs to be 1020 and not 120
{% elif p0>=120 and p0<1040 %}Q on p0… I am at 127m above sea level and 6km from the coast. At the start, being not sea-level I set this to 0 … if I set it to 1 then it will take my local pressure which is more in line to what I see. …will recommence testing
Just as information for other. The “short term forecast” is more a general weather condition. So high pressure area or low pressure area. For me the forecast still works…
My weather sensors produce pressure readings in inches of mercury, temperature in degrees Fahrenheit, and wind speed in miles per hour. What changes should I make in the code to accommodate? Thanks.
Perhaps changing your unit system in general settings
Had this working for 5 days, nothing happens to the card.
Please make a screencast how to set it up correctly.