summarize_networks InvalidIndexError #191
Conversation
| @@ -78,6 +78,7 @@ def summarize_networks(file_paths: Iterable[Union[str, PathLike]]) -> pd.DataFra | |||
| str(tup[0]): 1, | |||
| }, index=tup[1] | |||
| ) | |||
| dataframe = dataframe[~dataframe.index.duplicated(keep='first')] | |||
There was a problem hiding this comment.
There was a problem hiding this comment.
| dataframe = dataframe[~dataframe.index.duplicated(keep='first')] | |
| # Mark the first occurrence of a duplicate as True and all others as False | |
| # Non-duplicates are also marked as True | |
| dataframe = dataframe[~dataframe.index.duplicated(keep='first')] |
| @@ -78,6 +78,7 @@ def summarize_networks(file_paths: Iterable[Union[str, PathLike]]) -> pd.DataFra | |||
| str(tup[0]): 1, | |||
| }, index=tup[1] | |||
| ) | |||
| dataframe = dataframe[~dataframe.index.duplicated(keep='first')] | |||
There was a problem hiding this comment.
add a warning that edges are being dropped from a file
There was a problem hiding this comment.
If this is hard to do here, we can think about whether it should be handled somewhere else. For instance, should the parse_output functions check for duplicate edges and raise a warning or error?
There was a problem hiding this comment.
What you have looks good.
If you didn't already open an issue about MinCostFlow giving the duplicate issues that trigged this problem, please do that and link it to #188 (comment). We may want to follow up on that potential bug later.
| @@ -78,6 +78,7 @@ def summarize_networks(file_paths: Iterable[Union[str, PathLike]]) -> pd.DataFra | |||
| str(tup[0]): 1, | |||
| }, index=tup[1] | |||
| ) | |||
| dataframe = dataframe[~dataframe.index.duplicated(keep='first')] | |||
There was a problem hiding this comment.
| dataframe = dataframe[~dataframe.index.duplicated(keep='first')] | |
| # Mark the first occurrence of a duplicate as True and all others as False | |
| # Non-duplicates are also marked as True | |
| dataframe = dataframe[~dataframe.index.duplicated(keep='first')] |
| @@ -78,6 +78,7 @@ def summarize_networks(file_paths: Iterable[Union[str, PathLike]]) -> pd.DataFra | |||
| str(tup[0]): 1, | |||
| }, index=tup[1] | |||
| ) | |||
| dataframe = dataframe[~dataframe.index.duplicated(keep='first')] | |||
There was a problem hiding this comment.
If this is hard to do here, we can think about whether it should be handled somewhere else. For instance, should the parse_output functions check for duplicate edges and raise a warning or error?
PR for #188
Currently created a test case to showcase the error and how it is happening