We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Currently our slice definition looks as follows:
pub fn slice<D: std::ops::RangeBounds<usize>>(&self, ranges: &[D]) -> anyhow::Result<Tensor> { ///...impl... }
This is very user hostile, as the user must provide a homogeneous collection of ranges.
let y = x.slice(&[0..5, 0..6, 0..7]);
This is very annoying, as you may only care about one of the dimensions, it would be better to have something like
let y = x.slice(&[.., 0..6, ..]);
Unfortunately, this doesn't work, because the collection is now heterogeneous (i.e is made up of 2xRangeFull and 1xRange).
RangeFull
Range
Therefore, we need to use a macro, much like ndarray.
ndarray
let y = x.slice(s![.., 0..6, ..]);
This API gives the illusion of heterogeneous collections, which is what we want.
The text was updated successfully, but these errors were encountered:
No branches or pull requests
Currently our slice definition looks as follows:
This is very user hostile, as the user must provide a homogeneous collection of ranges.
This is very annoying, as you may only care about one of the dimensions, it would be better to have something like
Unfortunately, this doesn't work, because the collection is now heterogeneous (i.e is made up of 2x
RangeFull
and 1xRange
).Therefore, we need to use a macro, much like
ndarray
.This API gives the illusion of heterogeneous collections, which is what we want.
The text was updated successfully, but these errors were encountered: