Home My Page Projects Code Snippets Project Openings diderot

# SCM Repository

[diderot] Diff of /branches/ein16/TODO
 [diderot] / branches / ein16 / TODO

# Diff of /branches/ein16/TODO

revision 1155, Sun May 8 14:43:30 2011 UTC revision 1204, Thu May 12 19:10:20 2011 UTC
# Line 1  Line 1
1  NOTE: GLK's approximate ranking of 5 most important tagged with  NOTE: GLK's approximate ranking of 8 most important tagged with
2  [GLK:1], [GLK:2], ...  [GLK:1], [GLK:2], ...
3
4  ==============================  ========================
5  other SHORT TERM =============  (including needed for LIC)  SHORT TERM ============= (*needed* for streamlines & tractography)
6  ==============================  ========================

Add a clamp function, which takes three arguments; either three scalars:
clamp(x, minval, maxval)  = max(minval, min(maxval, x))
or three vectors of the same size:
clamp([x,y], minvec, maxvec)  = [max(minvec[0], min(maxvec[0], x)),
max(minvec[1], min(maxvec[1], y))]
This would be useful in many current Diderot programs.
One question: clamp(x, minval, maxval) is the argument order
used in OpenCL and other places, but clamp(minval, maxval, x)
would be more consistent with lerp(minout, maxout, x).
7
8  Level of differentiability in field type should be statement about how  [GLK:3] Add sequence types (needed for evals & evecs)
much differentiation the program *needs*, rather than what the kernel
*provides*.  The needed differentiability can be less than or equal to
the provided differentiability.

[GLK:1] Add sequence types (needed for evals & evecs)
9      syntax      syntax
10          types: ty '{' INT '}'          types: ty '{' INT '}'
11          value construction: '{' e1 ',' … ',' en '}'          value construction: '{' e1 ',' … ',' en '}'
12          indexing: e '{' e '}'          indexing: e '{' e '}'
13
14  IL support for higher-order tensor values (matrices, etc).  [GLK:4] evals & evecs for symmetric tensor[2,2] and
15      tensor construction [DONE]  tensor[3,3] (requires sequences)
tensor indexing [DONE]
tensor slicing
verify that hessians work correctly [DONE]
16
17  Use ∇⊗ etc. syntax  ability to emit/track/record variables into dynamically re-sized
18      syntax [DONE]  runtime buffer
typechecking
IL and codegen
19
20  test/uninit.diderot:  tensor fields: convolution on general tensor images
documents need for better compiler error messages when output variables
are not initialized; the current messages are very cryptic
21
22  determinant ("det") for tensor[3,3]  ========================
23    SHORT-ISH TERM ========= (to make using Diderot less annoying to
24    ========================  program in, and slow to execute)
25
26  expand trace in mid to low translation  value-numbering optimization [DONE, but needs more testing]
27
28  value-numbering optimization  Allow ".ddro" file extensions in addition to ".diderot"
29
30  Add type aliases for color types  Be able to output values of type tensor[2,2] and tensor[3,3];
31      rgb = real{3}  (currently only scalars & vectors).  Want to add some regression tests
32      rgba = real{4}  based on this and currently can't
33
34  ==============================  [GLK:1] Add a clamp function, which takes three arguments; either
35  MEDIUM TERM ================== (including needed for streamlines & tractography)  three scalars:
36  ==============================    clamp(lo, hi, x)  = max(lo, min(hi, x))
37    or three vectors of the same size:
38      clamp(lo, hi, [x,y])  = [max(lo[0], min(hi[0], x)),
39                               max(lo[1], min(hi[1], y))]
40    This would be useful in many current Diderot programs.
41    One question: clamp(x, lo, hi) is the argument order used in OpenCL
42    and other places, but clamp(lo, hi, x) is much more consistent with
43    lerp(lo, hi, x), hence GLK's preference
44
45  [GLK:1] evals & evecs for symmetric tensor[3,3] (requires sequences)  [GLK:2] Proper handling of stabilize method
46
47  [GLK:2] Save Diderot output to nrrd, instead of "mip.txt"  allow "*" to represent "modulate": per-component multiplication of
48    vectors, and vectors only (not tensors of order 2 or higher).  Once
49    sequences are implemented this should be removed: the operation is not
50    invariant WRT basis so it is not a legit vector computation.
51
52    implicit type promotion of integers to reals where reals are
53    required (e.g. not exponentiation "^")
54
55    [GLK:5] Save Diderot output to nrrd, instead of "mip.txt"
56    For grid of strands, save to similarly-shaped array    For grid of strands, save to similarly-shaped array
57    For list of strands, save to long 1-D (or 2-D for non-scalar output) list    For list of strands, save to long 1-D (or 2-D for non-scalar output) list
58    For ragged things (like tractography output), will need to save both    For ragged things (like tractography output), will need to save both
59      complete list of values, as well as list of start indices and lengths      complete list of values, as well as list of start indices and lengths
60      to index into complete list      to index into complete list
61
62  [GLK:3] Use of Teem's "hest" command-line parser for getting  [GLK:6] Use of Teem's "hest" command-line parser for getting
63  any input variables that are not defined in the source file  any input variables that are not defined in the source file
64
65  [GLK:4] ability to declare a field so that probe positions are  [GLK:7] ability to declare a field so that probe positions are
66  *always* "inside"; with various ways of mapping the known image values  *always* "inside"; with various ways of mapping the known image values
67  to non-existant index locations.  One possible syntax emphasizes that  to non-existant index locations.  One possible syntax emphasizes that
68  there is a index mapping function that logically precedes convolution:  there is a index mapping function that logically precedes convolution:
# Line 76  Line 71
71    F = bspln3 ⊛ (img ◦ mirror)    F = bspln3 ⊛ (img ◦ mirror)
72  where "◦" or "∘" is used to indicate function composition  where "◦" or "∘" is used to indicate function composition
73
74  extend norm (|exp|) to all tensor types [DONE for vectors and matrices]  Level of differentiability in field type should be statement about how
75    much differentiation the program *needs*, rather than what the kernel
76  ability to emit/track/record variables into dynamically re-sized  *provides*.  The needed differentiability can be less than or equal to
77  runtime buffer  the provided differentiability.

Want: allow X *= Y, X /= Y, X += Y, X -= Y to mean what they do in C,
provided that X*Y, X/Y, X+Y, X-Y are already supported.
Nearly every Diderot program would be simplified by this.
78
79  [GLK:5] Want: non-trivial field expressions & functions:  Use ∇⊗ etc. syntax
80    image(2)[2] Vimg = load(...);      syntax [DONE]
81    field#0(2)[] Vlen = |Vimg ⊛ bspln3|;      typechecking
82  to get a scalar field of vector length, or      IL and codegen
field#2(2)[] F = Fimg ⊛ bspln3;
field#0(2)[] Gmag = |∇F|;
to get a scalar field of gradient magnitude, or
field#2(2)[] F = Fimg ⊛ bspln3;
field#0(2)[] Gmsq = ∇F•∇F;
to get a scalar field of squared gradient magnitude, which is simpler
to differentiate.  However, there is value in having these, even if
the differentiation of them is not supported (hence the indication
of "field#0" for these above)

Want: ability to apply "normalize" to a field itself, e.g.
field#0(2)[2] V = normalize(Vimg ⊛ ctmr);
so that V(x) = normalize((Vimg ⊛ ctmr)(x)).
Having this would simplify expression of standard LIC method, and
would also help express other vector field expressions that arise
in vector field feature exraction.
83
84  tensor fields: convolution on general tensor images  Add type aliases for color types
85        rgb = real{3}
86        rgba = real{4}
87
88  ==============================  ==============================
89  other MEDIUM TERM ============ (needed for particles)  MEDIUM TERM ================== (*needed* for particles)
90  ==============================  ==============================
91
Put small 1-D and 2-D fields, when reconstructed specifically by tent
and when differentiation is not needed, into faster texture buffers.
test/illust-vr.diderot is good example of program that uses multiple
such 1-D fields basically as lookup-table-based function evaluation

92  run-time birth of strands  run-time birth of strands
93
94  "initially" supports lists  "initially" supports lists
# Line 124  Line 96
96  "initially" supports lists of positions output from  "initially" supports lists of positions output from
97  different initalization Diderot program  different initalization Diderot program
98
99  spatial data structure that permits strands' queries of neighbors  Communication between strands: they have to be able to learn each
100    other's state (at the previous iteration).  Early version of this can
101    have the network of neighbors be completely static (for running one
102    strand/pixel image computations).  Later version with strands moving
103    through the domain will require some spatial data structure to
104    optimize discovery of neighbors.
105
106    ============================
107    MEDIUM-ISH TERM ============ (to make Diderot more useful/effective)
108    ============================
109
110  proper handling of stabilize method  Python/ctypes interface to run-time
111
112  test/vr-kcomp2.diderot: Add support for code like  support for Python interop and GUI
113
114          (F1 if x else F2)@pos  Allow integer exponentiation ("^2") to apply to square matrices,
115    to represent repeated matrix multiplication
116
117  This will require duplication of the continuation of the conditional  Alow X *= Y, X /= Y, X += Y, X -= Y to mean what they do in C,
118  (but we should only duplicate over the live-range of the result of the  provided that X*Y, X/Y, X+Y, X-Y are already supported.
119  conditional.  Nearly every Diderot program would be simplified by this.
120
121    Put small 1-D and 2-D fields, when reconstructed specifically by tent
122    and when differentiation is not needed, into faster texture buffers.
123    test/illust-vr.diderot is good example of program that uses multiple
124    such 1-D fields basically as lookup-table-based function evaluation
125
126    expand trace in mid to low translation
127
128    extend norm (|exp|) to all tensor types [DONE for vectors and matrices]
129
130    determinant ("det") for tensor[3,3]
131
132  add ":" for tensor dot product (contracts out two indices  add ":" for tensor dot product (contracts out two indices
133  instead of one like •), valid for all pairs of tensors with  instead of one like •), valid for all pairs of tensors with
134  at least two indices  at least two indices
135
136  ==============================  test/uninit.diderot:
137  other MEDIUM TERM ============  documents need for better compiler error messages when output variables
138  ==============================  are not initialized; the current messages are very cryptic
139
140  want: warnings when "D" (reserved for differentiation) is declared as  want: warnings when "D" (reserved for differentiation) is declared as
141  a variable name (get confusing error messages now)  a variable name (get confusing error messages now)
142
143  support for Python interop and GUI  ==============================
144    LONG TERM ==================== (make Diderot more interesting/attractive from
145    ==============================  a research standpoint)
146
147  Python/ctypes interface to run-time  IL support for higher-order tensor values (matrices, etc).
148        tensor construction [DONE]
149        tensor indexing [DONE]
150        tensor slicing
151        verify that hessians work correctly [DONE]
152
153    Better handling of variables that determines the scope of a variable
154    based on its actual use, instead of where the user defined it.  So,
155    for example, we should lift strand-invariant variables to global
156    scope.  Also prune out useless variables, which should include field
157    variables after the translation to mid-il.
158
159    test/vr-kcomp2.diderot: Add support for code like
160            (F1 if x else F2)@pos
161    This will require duplication of the continuation of the conditional
162    (but we should only duplicate over the live-range of the result of the
163    conditional.
164
165    [GLK:8] Want: non-trivial field expressions & functions.
166    scalar fields from scalar fields F and G:
167      field#0(2)[] X = (sin(F) + 1.0)/2;
168      field#0(2)[] X = F*G;
169    scalar field of vector field magnitude:
171      field#0(2)[] Vlen = |Vimg ⊛ bspln3|;
172    field of normalized vectors (for LIC and vector field feature extraction)
173      field#2(2)[2] F = ...
174      field#0(2)[2] V = normalize(F);
175    scalar field of gradient magnitude (for edge detection))
176      field#2(2)[] F = Fimg ⊛ bspln3;
177      field#0(2)[] Gmag = |∇F|;
178    scalar field of squared gradient magnitude (simpler to differentiate):
179      field#2(2)[] F = Fimg ⊛ bspln3;
180      field#0(2)[] Gmsq = ∇F•∇F;
181    There is value in having these, even if the differentiation of them is
182    not supported (hence the indication of "field#0" for these above)
183
184    Introduce region types (syntax region(d), where d is the dimension of the
185    region.  One useful operator would be
186            dom : field#k(d)[s] -> region(d)
187    Then the inside test could be written as
188            pos ∈ dom(F)
189    We could further extend this approach to allow geometric definitions of
190    regions.  It might also be useful to do inside tests in world space,
192
193    co- vs contra- index distinction
194
195    Permit field composition:
196      field#2(3)[3] warp = bspln3 ⊛ warpData;
197      field#2(3)[] F = bspln3 ⊛ img;
198      field#2(3)[] Fwarp = F ◦ warp;
199    So Fwarp(x) = F(warp(X)).  Chain rule can be used for differentation.
200    This will be instrumental for expressing non-rigid registration
201    methods (but those will require co-vs-contra index distinction)
202
203  Allow the convolution to be specified either as a single 1D kernel  Allow the convolution to be specified either as a single 1D kernel
204  (as we have it now):  (as we have it now):
# Line 158  Line 207
207    field#0(3)[] F = (bspln3 ⊗ bspln3 ⊗ tent) ⊛ img;    field#0(3)[] F = (bspln3 ⊗ bspln3 ⊗ tent) ⊛ img;
208  This is especially important for things like time-varying data, or  This is especially important for things like time-varying data, or
209  other multi-dimensional fields where one axis of the domain is very  other multi-dimensional fields where one axis of the domain is very
210  different from the rest.  What is very unclear is how, in such cases,  different from the rest, and hence must be treated separately when
211    it comes to convolution.  What is very unclear is how, in such cases,
212  we should notate the gradient, when we only want to differentiate with  we should notate the gradient, when we only want to differentiate with
213  respect to some of the axes.  respect to some subset of the axes.  One ambitious idea would be:
214      field#0(3)[] Ft = (bspln3 ⊗ bspln3 ⊗ tent) ⊛ img; // 2D time-varying field
215      field#0(2)[] F = lambda([x,y], Ft([x,y,42.0]))     // restriction to time=42.0
217
218    Tensors of order 3 (e.g. gradients of diffusion tensor fields, or
219    hessians of vector fields) and order 4 (e.g. Hessians of diffusion
220    tensor fields).
221
222  ==============================  representation of tensor symmetry
LONG TERM ====================
==============================

Better handling of variables that determines the scope of a variable
based on its actual use, instead of where the user defined it.  So,
for example, we should lift strand-invariant variables to global
scope.  Also prune out useless variables, which should include field
variables after the translation to mid-il.

co- vs contra- index distinction

some indication of tensor symmetry
223  (have to identify the group of index permutations that are symmetries)  (have to identify the group of index permutations that are symmetries)
224
225  dot works on all tensors  dot works on all tensors
226
227  outer works on all tensors  outer works on all tensors
228
229    Help for debugging Diderot programs: need to be able to uniquely
230    identify strands, and for particular strands that are known to behave
231    badly, do something like printf or other logging of their computations
233
234    Permit writing dimensionally general code: Have some statement of the
235    dimension of the world "W" (or have it be learned from one particular
236    field of interest), and then able to write "vec" instead of
237    "vec2/vec3", and perhaps "tensor[W,W]" instead of
238    "tensor[2,2]/tensor[3,3]"
239
240    Traits: all things things that have boilerplate code (especially
241    volume rendering) should be expressed in terms of the unique
242    computational core.  Different kinds of streamline/tractography
243    computation will be another example, as well as particle systems.
244
245  Einstein summation notation  Einstein summation notation
246
247  "tensor comprehension" (like list comprehension)  "tensor comprehension" (like list comprehension)
248
249    Fields coming from different sources of data:
250    * triangular or tetrahedral meshes over 2D or 3D domains (of the
251      source produced by finite-element codes; these will come with their
252      own specialized kinds of reconstruction kernels, called "basis
253      functions" in this context)
254    * Large point clouds, with some radial basis function around each point,
255      which will be tuned by parameters of the point (at least one parameter
256      giving some notion of radius)
257
258  ======================  ======================
259  BUGS =================  BUGS =================
260  ======================  ======================

Legend:
 Removed from v.1155 changed lines Added in v.1204