[issue-1248] features inference for passed into check dataframes

[yromanyshyn]

resolves #1248

@Nadav-Barak
it is enough to modify line 889 in Dataset.cast_to_dataset to solve that issue, but I am not sure how clear the warning message raised by the features inference logic (Dataset._infer_categorical_features line 573) will be for the users in this case. What do you think?

[matanper]

I think that in Dataset line 889 we can add the warning "DataFrame was passed, initializing deepchecks.tabular.Dataset using the given dataframe"

[Nadav-Barak]

I think the warning massage is good for both cases (if the user passed a DF directly or created a DS without specifying categorical)

[yromanyshyn]

@matanper

I think that in Dataset line 889 we can add the warning "DataFrame was passed, initializing deepchecks.tabular.Dataset using the given dataframe"

Think it will be redundant, this function is part of a public API and it could be strange for user to see warning as this when he/she will try to use this function directly

[matanper]

@matanper

I think that in Dataset line 889 we can add the warning "DataFrame was passed, initializing deepchecks.tabular.Dataset using the given dataframe"

Think it will be redundant, this function is part of a public API and it could be strange for user to see warning as this when he/she will try to use this function directly

I don't see a reason for it to be a public API, it doesn't help the user more than calling the ctor

[yromanyshyn]

@matanper

I don't see a reason for it to be a public API, it doesn't help the user more than calling the ctor

ok, I have not changed its visibility marker, but I have added a warning