Python Script stopping at 30%

I’m a new Knime user (V 3.7.1) coming from a Python programming background. I have installed Python 3.6 using Anaconda on Windows 10. I am munging data using Python Script (1=>1). I have tested my script in GradientBoostingRegressor and it works. I have replicated and executed the script within the Python Script node and it runs fine. I added a print statement at the end of my script to check, and my code generates the output I am looking for. My problem comes when I close the node and then try to execute it - it remains at 30% and eventually fails. Can anyone explain to why this is happening and how I can resolve it. Its incredibly frustrating

import pandas as pd
import numpy as np
from sklearn.ensemble import GradientBoostingRegressor




def log_transform(feature):
train[feature] = np.log1p(train[feature].values)

def quadratic(feature):
train[feature + ‘2’] = train[feature] ** 2

gbm = GradientBoostingRegressor(n_estimators=4000, alpha=0.01);
y_train_log = np.log1p(Y), y_train_log)
preds = gbm.predict(X)


predicted_values = np.exp(y_train_log)
actual_values = np.exp(preds)

predicted_values = np.array(predicted_values)
actual_values = np.array(actual_values)

log_predict = np.log(predicted_values + 1)
log_actual = np.log(actual_values + 1)

difference = log_predict - log_actual
difference = np.square(difference)

mean_difference = difference.mean()

score = np.sqrt(mean_difference)

#print ("RMSLE Value For Gradient Boost: ", score)

result = pd.DataFrame(np.exp(preds))

The Python nodes in KNIME follow a strict input and output system. In a ‘normal’ Python script you would have one or two data tables that would come out as a result, or you would have a model or a graphic. Of course you could send a path to the Python script and then just store everything you want directly from the Python script on your hard drive, but that is not the KNIME-like way.

Here you have a few examples of KNIME and Python nodes:

1 Like

This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.