❌ Execution failed: Could not connect to the Python process.

:x: Execution failed: Could not connect to the Python process.
i run a workflow in knime, but i can’t, this is my script

import knime.scripting.io as knio

Import required libraries

import pandas as pd

import numpy as np

Read the input data into a pandas DataFrame

data = knio.input_tables[0].to_pandas()

Define the input sequence length and number of future predictions

num_predictions = max(data[‘date_range’].values)

Create a date range

start_date = min(data[‘Min(date_order)’].values)

end_date = max(data[‘Max(date_order)’].values)

date_range = pd.date_range(start_date + pd.DateOffset(0), periods=num_predictions, freq=‘D’)

Create a dataframe with the predicted values and corresponding dates

predicted_data = pd.DataFrame({‘date_knime’: date_range})

Export the predicted data to output.table

knio.output_tables[0] = knio.input_tables[0]

knio.output_tables[1] = knio.Table.from_pandas(predicted_data)

whereas, i have installed, numpy, pandas and also prophet using conda prompt.

Hi @Gelora_Fikrisin ,
If you are using your own Python environment, you may need to check whether the “knime-python-scripting” package has been installed in the current environment.

In the self-installed Python environment, this package needs to be installed additionally.

See the document in detail:
https://docs.knime.com/latest/python_installation_guide/index.html#_configure_the_ap_wide_environment

5 Likes

In addition to the guide I’ve also written an article on medium on this topic (if you know and like the game snake this may also be for you ;-)):

3 Likes

Very detailed operation guide! :+1: :smiley:

1 Like