Closed LittleLittleCloud closed 4 years ago
@LittleLittleCloud can you add more details?
@LittleLittleCloud I'm assuming this is a dataset with a weird header. Can you share the dataset?
@LittleLittleCloud I'm assuming this is a dataset with a weird header. Can you share the dataset?
It's not a dataset problem, going to move it from Dataset Failure
epic
@LittleLittleCloud, @briacht Can we update this message to be more clear? This is the error message we currently get when has-header was set to "false" but we detect a header:
"Detect result for input file is True while your input is False. Please double check to confirm."
Does the user have to input something here if it's just a warning message? Or what are the actions they are supposed to take?
"A header has been detected in your dataset, but you have specified --has-header false. Are you sure that your dataset does not have a header? (Input y/n)."
No it's just a warning msg and they can't do any thing. So maybe just
A header has been detected in your dataset, but you have specified --has-header false.
when has header
and
There's no header been detected in your dataset, but you have specified --has-header true.
?
Does this look like too much text in the CLI:
Warning: A header has been detected in your dataset, but you have specified --has-header false. If results are not as expected, please check --has-header option and try training again.
Warning: No header has been detected in your dataset, but you have specified --has-header true. If results are not as expected, please check --has-header option and try training again.
It's OK, text message is never too much in cli program
mlnet-cli should show warning message if --has-header flag is contradicted with auto-detected result.