Time Series Forecasting with fbProphet

#Timeseries analysis involves examining data at specific times to identify trends and patterns. @deganza develops a #lowcode solution with #KNIME to #forecast #airline passengers and the #weather using the #Facebook #Prophet algorithm. Enjoy the data story!

PS: :date:#HELPLINE . Want to discuss your article? Need help structuring your story? Make a date with the editors of Low Code for Data Science via Calendly → Calendly - Blog Writer

2 Likes

Followed the installation steps and got error message:
Execution failed: Could not connect to the Python process
I’ve tested other workflows with the Python Script node and they work.

Hi @rfeigel, thanks for the heads-up. I haven’t seen this error myself but another user had this issue before and a solution was suggested (see below)

Could you maybe try this:

Hope it helps,
Roberto

I’m thoroughly confused. I tried to point Knime directly to the Python Base environment, but its not available in the env folder. I installed Prophet and the latest Pandas in another environment and tried to run that. The Python Script node now throws the error below when executing the script line by line.




Hi @rfeigel,

Sorry I must have missed your message. Did you manage to find a solution in the meantime?

If not, please check this answer: ❌ Execution failed: Could not connect to the Python process. - #2 by tomljh. See if it helps!

Roberto

Sorry for the trouble. Here’s the error message I’m getting:
Execute failed: RuntimeError: Error during optimization! Command ‘C:\Users\Owner\anaconda3\envs\py39\Lib\site-packages\prophet\stan_model\prophet_model.bin random seed=94065 data file=C:\Users\Owner\AppData\Local\Temp\tmpprr26dy3\fn7j4pjm.json init=C:\Users\Owner\AppData\Local\Temp\tmpprr26dy3\14lvz8d5.json output file=C:\Users\Owner\AppData\Local\Temp\tmpprr26dy3\prophet_model9jcgr4o6\prophet_model-20240712113635.csv method=optimize algorithm=newton iter=10000’ failed:

Hi @rfeigel,

unfortunately, I don’t understand what may cause the error that you get.

I’ve added a Conda Environment Propagation node to the original workflow so hopefully you should be able to recreate my fbprophet environment on your machine (I’m on Win) and run the Python Script nodes successfully.

Hope this helps!
Roberto

fbProphet_Component.knwf (168.0 KB)

1 Like

Thanks so much for taking the time and trouble to help me. Works perfectly. Created the fbprophet env. The error messages I got were so obscure that I couldn’t understand what was wrong with the environment I was trying to use. I’m a Python novice at best.

1 Like