Resolving the "KernelNotResponding: The kernel died unexpectedly" error
Issue description
When working with CSV files, you see this error:
KernelNotResponding: The kernel died unexpectedly
Solution
Try reading the first several dozen lines from a CSV file. Use pandas.read_csv and specify the additional nrows parameter to determine the number of necessary rows in the nrows format.
You can also read only relevant columns from the CSV file: index_col or names.
After loading the fragment, you can set the footprint in RAM like this: df.memory_usage(deep=True).sum(). This will help you define the relevant node configuration.
For details about the RAM amount per configuration, see our documentation.
Note
Access to configurations starting from g1.1 becomes available to DataSphere on switching to paid usage and topping up the account by at least 500 rubles.
After switching to paid usage, any active grants will be applied first towards payment for resources. We will only start debiting your account after the grant is used up or has expired.