Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

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

Jump to bottom

Add indexing of DiscreteLp #1113

Open
adler-j opened this issue Sep 6, 2017 · 2 comments
Open

Add indexing of DiscreteLp #1113

adler-j opened this issue Sep 6, 2017 · 2 comments
Labels
area: core status: discussion type: new feature

Comments

@adler-j
Copy link
Member

adler-j commented Sep 6, 2017

DiscreteLp should be indexable like geometries, according to the rule:

space[idx].partition == space.partition[idx]
@kohr-h
Copy link
Member

kohr-h commented Sep 6, 2017

I see an upcoming issue here with vector- and tensor-valued functions. If we have a vector-valued function with 2 components (say a gradient in 2D), wouldn't we want f[0] be the first component and f[1] the second, in contrast to indexing the domain?

In the face of this ambiguity it may be better to not implement __getitem__ and instead have two properties, e.g., slice_domain and slice_range or so, similar to byaxis. If over time there turns out to be a clear preference we can still make __getitem__ an alias for that preference.

@adler-j
Copy link
Member Author

adler-j commented Sep 6, 2017 •
edited
Loading

Well I see no conflict here, one would simply expect a tensor valued function (say with rank n over a k dimensional space) to be indexed like

x[i1, ..., in, i{n+1}, ..., i{n+k}]
   ^ "tensor"   ^ "partition"

I guess this breaks my above code, but only in an obvious way.

With that said adding your utilities would certainly be good.

@adler-j adler-j mentioned this issue Nov 24, 2017
Closed
kohr-h pushed a commit to kohr-h/odl that referenced this issue Dec 11, 2017
ENH: implement tensor-valued DiscreteLp
53d02c4 
Changes in detail:
- Add dtype with shape to DiscreteLp (mostly __repr__, factory
  functions and some downstream methods). As a consequence,
  `shape_[in,out]` and `ndim_[in,out]` are added for the
  different types of axes, as well as `scalar_dtype`.
- Add `PerAxisWeighting` and make it the default for
  `DiscreteLp` type spaces. Reason: this way the `tspace`
  knows how to deal with removed axes etc. This is important
  for a smooth experience with indexing and reductions over
  axes.
  Helpers for slicing weightings help structure this task.
- Implement `__getitem__` for `TensorSpace` and `DiscreteLp`,
  including (hopefully) reasonable propagation of weights.
  The new `simulate_slicing` utility function simplifies
  this task.
- Allow indexing with ODL tensors of boolean or integer dtype.
- Implement correct weighting for backprojections with
  non-uniform angles, using per-axis weighting and a new
  helper `adjoint_weightings` to apply the weightings in an
  efficient way.
  The correct weighting from the range of `RayTransform`
  is determined by the new `proj_space_weighting` helper.
- Change the space `_*_impl` methods to always expect and
  return Numpy arrays, and adapt the calling code.
- Change behavior of `norm` and `dist` to ignoring weights
  for `exponent=inf`, in accordance with math.
- Improve speed of `all_equal` for comparison of arrays.
- Account for `None` entries in indices in the
  `normalized_index_expression` helper, thus allowing
  creation of new axes.
- Remove `dicsr_sequence_space`, it was largely unused and
  just a maintenance burden. Use a regular `uniform-discr`
  from zero to `shape` instead.
- Remove `Weighting.equiv()` mehtods, never used and hard
  to maintain (n^2 possibilities).
- Remove the (largely useless) `_weighting` helper to create
  weighting instances since it would have been ambiguous
  with sequences of scalars (array or per axis?). Also remove
  the `npy_weighted_*` functions, they were useless, too.
- Remove some dead code from tomo/util.
- A bunch of minor fixes, as usual.

Closes: odlgroup#908, odlgroup#907, odlgroup#1113, odlgroup#965, odlgroup#286, odlgroup#267, odlgroup#1001
kohr-h pushed a commit to kohr-h/odl that referenced this issue Dec 11, 2017
ENH: implement tensor-valued DiscreteLp
bb31079 
Changes in detail:
- Add dtype with shape to DiscreteLp (mostly __repr__, factory
  functions and some downstream methods). As a consequence,
  `shape_[in,out]` and `ndim_[in,out]` are added for the
  different types of axes, as well as `scalar_dtype`.
- Add `PerAxisWeighting` and make it the default for
  `DiscreteLp` type spaces. Reason: this way the `tspace`
  knows how to deal with removed axes etc. This is important
  for a smooth experience with indexing and reductions over
  axes.
  Helpers for slicing weightings help structure this task.
- Implement `__getitem__` for `TensorSpace` and `DiscreteLp`,
  including (hopefully) reasonable propagation of weights.
  The new `simulate_slicing` utility function simplifies
  this task.
- Allow indexing with ODL tensors of boolean or integer dtype.
- Implement correct weighting for backprojections with
  non-uniform angles, using per-axis weighting and a new
  helper `adjoint_weightings` to apply the weightings in an
  efficient way.
  The correct weighting from the range of `RayTransform`
  is determined by the new `proj_space_weighting` helper.
- Change the space `_*_impl` methods to always expect and
  return Numpy arrays, and adapt the calling code.
- Change behavior of `norm` and `dist` to ignoring weights
  for `exponent=inf`, in accordance with math.
- Improve speed of `all_equal` for comparison of arrays.
- Account for `None` entries in indices in the
  `normalized_index_expression` helper, thus allowing
  creation of new axes.
- Remove `dicsr_sequence_space`, it was largely unused and
  just a maintenance burden. Use a regular `uniform-discr`
  from zero to `shape` instead.
- Remove `Weighting.equiv()` mehtods, never used and hard
  to maintain (n^2 possibilities).
- Remove the (largely useless) `_weighting` helper to create
  weighting instances since it would have been ambiguous
  with sequences of scalars (array or per axis?). Also remove
  the `npy_weighted_*` functions, they were useless, too.
- Remove some dead code from tomo/util.
- A bunch of minor fixes, as usual.

Closes: odlgroup#908
Closes: odlgroup#907
Closes: odlgroup#1113
Closes: odlgroup#965
Closes: odlgroup#286
Closes: odlgroup#267
Closes: odlgroup#1001
kohr-h pushed a commit to kohr-h/odl that referenced this issue Jun 30, 2018
@kohr-h
ENH: implement tensor-valued DiscreteLp
2d7fc20 
Changes in detail:
- Add dtype with shape to DiscreteLp (mostly __repr__, factory
  functions and some downstream methods). As a consequence,
  `shape_[in,out]` and `ndim_[in,out]` are added for the
  different types of axes, as well as `scalar_dtype`.
- Add `PerAxisWeighting` and make it the default for
  `DiscreteLp` type spaces. Reason: this way the `tspace`
  knows how to deal with removed axes etc. This is important
  for a smooth experience with indexing and reductions over
  axes.
  Helpers for slicing weightings help structure this task.
- Implement `__getitem__` for `TensorSpace` and `DiscreteLp`,
  including (hopefully) reasonable propagation of weights.
  The new `simulate_slicing` utility function simplifies
  this task.
- Allow indexing with ODL tensors of boolean or integer dtype.
- Implement correct weighting for backprojections with
  non-uniform angles, using per-axis weighting and a new
  helper `adjoint_weightings` to apply the weightings in an
  efficient way.
  The correct weighting from the range of `RayTransform`
  is determined by the new `proj_space_weighting` helper.
- Change the space `_*_impl` methods to always expect and
  return Numpy arrays, and adapt the calling code.
- Change behavior of `norm` and `dist` to ignoring weights
  for `exponent=inf`, in accordance with math.
- Improve speed of `all_equal` for comparison of arrays.
- Account for `None` entries in indices in the
  `normalized_index_expression` helper, thus allowing
  creation of new axes.
- Remove `dicsr_sequence_space`, it was largely unused and
  just a maintenance burden. Use a regular `uniform-discr`
  from zero to `shape` instead.
- Remove `Weighting.equiv()` mehtods, never used and hard
  to maintain (n^2 possibilities).
- Remove the (largely useless) `_weighting` helper to create
  weighting instances since it would have been ambiguous
  with sequences of scalars (array or per axis?). Also remove
  the `npy_weighted_*` functions, they were useless, too.
- Remove some dead code from tomo/util.
- A bunch of minor fixes, as usual.

Closes: odlgroup#908
Closes: odlgroup#907
Closes: odlgroup#1113
Closes: odlgroup#965
Closes: odlgroup#286
Closes: odlgroup#267
Closes: odlgroup#1001
kohr-h pushed a commit to kohr-h/odl that referenced this issue Sep 12, 2018
@kohr-h
ENH: implement tensor-valued DiscreteLp
41b1d8e 
Changes in detail:
- Add dtype with shape to DiscreteLp (mostly __repr__, factory
  functions and some downstream methods). As a consequence,
  `shape_[in,out]` and `ndim_[in,out]` are added for the
  different types of axes, as well as `scalar_dtype`.
- Add `PerAxisWeighting` and make it the default for
  `DiscreteLp` type spaces. Reason: this way the `tspace`
  knows how to deal with removed axes etc. This is important
  for a smooth experience with indexing and reductions over
  axes.
  Helpers for slicing weightings help structure this task.
- Implement `__getitem__` for `TensorSpace` and `DiscreteLp`,
  including (hopefully) reasonable propagation of weights.
  The new `simulate_slicing` utility function simplifies
  this task.
- Allow indexing with ODL tensors of boolean or integer dtype.
- Implement correct weighting for backprojections with
  non-uniform angles, using per-axis weighting and a new
  helper `adjoint_weightings` to apply the weightings in an
  efficient way.
  The correct weighting from the range of `RayTransform`
  is determined by the new `proj_space_weighting` helper.
- Change the space `_*_impl` methods to always expect and
  return Numpy arrays, and adapt the calling code.
- Change behavior of `norm` and `dist` to ignoring weights
  for `exponent=inf`, in accordance with math.
- Improve speed of `all_equal` for comparison of arrays.
- Account for `None` entries in indices in the
  `normalized_index_expression` helper, thus allowing
  creation of new axes.
- Remove `dicsr_sequence_space`, it was largely unused and
  just a maintenance burden. Use a regular `uniform-discr`
  from zero to `shape` instead.
- Remove `Weighting.equiv()` mehtods, never used and hard
  to maintain (n^2 possibilities).
- Remove the (largely useless) `_weighting` helper to create
  weighting instances since it would have been ambiguous
  with sequences of scalars (array or per axis?). Also remove
  the `npy_weighted_*` functions, they were useless, too.
- Remove some dead code from tomo/util.
- A bunch of minor fixes, as usual.

Closes: odlgroup#908
Closes: odlgroup#907
Closes: odlgroup#1113
Closes: odlgroup#965
Closes: odlgroup#286
Closes: odlgroup#267
Closes: odlgroup#1001
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
area: core status: discussion type: new feature
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@adler-j @kohr-h