I want to forecast the hot spot temperature of power transformer using the Multivariate Time Series Analysis: LSTMs & Codeless. I have followed the preprocessing steps and it worked fine. However, when I have reached the step of closing the loop by using the loop close append it gives me an error and says that one of the input columns does not exist. The column existed in the previous steps, only when I reached to this point the column name is changed or disappear and I get this message "WARN Lag Column 6:11 Selected column “HUFL” does not exist.
Note: my data set contain six variables as an input and one variable as the target which I want to forecast. The samples have been collected every hour in one year.
I do not really know what the problem with this node is I have tried many times but I could not.
Is there any alternatives for it? or what is the solution
thank you for your prompt reply. I have fixed this problem by copying the blocks from the original workflow and it worked fine. However, I came with another problem. I tried to run keras node learner by initializing its parameters, but I got another error stating the following:
"ERROR Keras Network Learner 6:194 Execute failed: An error occurred while creating the Keras network from its layer specifications. Details:
An error occurred while trying to launch Python: Could not start Python. There are problems with your Python environment:
Could not start Python executable at the given location (no_conda_environment_selected\python.exe): Cannot run program “no_conda_environment_selected\python.exe”: CreateProcess error=2, The system cannot find the file specified
In case Python packages are missing you can create a new Python Conda environment that contains all packages required by the KNIME deep learning integrations in the “KNIME Deep Learning” Preferences.
How can I integrate knime with Python conda environment?
Can I install the python packages directly to knime?
KNIME_project2.knwf (91.6 KB)
Hi normally you set the environment in the preferences section.
Packages are installed in your environment not directly in KNIME.
You can share your python packages (e.g the environment setup) using conda propagation node with others when sharing your workflow.
br
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.