Filtering NaN Results in Extracted Trading Features
Summary
This brief platform support exchange distinguishes missing feature values from NaN values produced by calculations. The support response says missing values had already been filtered; remaining NaNs were calculation results, such as an undefined zero divided by zero. It suggests filtering the affected column by comparing it with the string representation of NaN.
The user then asks how to apply that filter across ten features at once, or whether there is a general way to handle not-a-number results. The exchange does not provide an answer to this multi-column question, so it offers a useful distinction but no complete implementation method. It also gives no examples of how the platform represents NaN across data types or how filtering affects downstream analysis.
Key ideas
- Missing feature values and calculation-generated NaNs may be treated differently by the platform.
- A zero divided by zero is given as an example of a calculation that produces NaN.
- The support response suggests filtering a column by comparing it with a NaN string representation.
- The discussion leaves bulk filtering across multiple features unresolved.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.