[WIP] Using c++ templates #18

kvmanohar22 · 2017-03-03T23:05:56Z

In Ref #16
Adding functionality of templates. Will be updated with the usage of pointers for faster algorithm.

maxruby · 2017-03-04T17:27:24Z

Nice to see such a quick start! Please see my comments in #17. Please remember to write tests to demonstrate that these new methods indeed work with Templates.

kvmanohar22 · 2017-03-04T21:58:33Z

I have added the test cases but couldn't test it locally on my laptop. How about we get travis up and running ?

maxruby · 2017-03-04T22:03:33Z

Sure, we can get Travis CI running, but you and I should definitely be able to test everything locally first. I am looking into the error you reported with dlopen_e in #17. I think I have seen this before.

maxruby · 2017-03-05T09:42:42Z

I have now built the latest julia-0.6.0 master, tested Cxx.jl and OpenCV.jl, solved several issues relevant to moving this forward. I will push the commits today and then you can work with the latest version of OpenCV.jl.

In the meantime, I wanted to note that I had already implemented templates for std::vector containers in OpenCV.jl and used them effectively too :) See OpenCV_util.jl

# C++ std::vector class thin-wrappings for Julia
cxx"""
template <typename T>
    std::vector<T> stdvector(int size, T x)
      {
        std::vector<T> cppvec(size, x);
        return (cppvec);
      }
. . .
"""

maxruby

This statement won't be reached. There is conditional statement any longer in this function.

maxruby · 2017-03-05T13:00:54Z

src/OpenCV_Mat.jl

+  return (val)
+
+  # RGB images
+  vec = @cxx ptr_val3(img, row, col);


This statement will not be reached because there is no conditional statement anymore in the function.

Oops ! I never considered that. Made the changes.

maxruby · 2017-03-06T23:03:48Z

src/OpenCV_Mat.jl

-  vec = @cxx ptr_val3(img, row, col);
-  println([int(at(vec, 0)), int(at(vec, 1)), int(at(vec, 2))])
-  return (vec)
+  image = Images.separate(img);


We should generally avoid more dependencies than absolutely necessary (e.g., Images), especially when there are already methods in OpenCV.jl that can achieve the same goal.

The following lines:

image = Images.separate(img); cd = Images.colordim(image)

can be a single line as follows:

channels(img)

Points worth keeping in mind:

separate and colordim are deprecated

ImageCore.jl is a much smaller dependency

In modern Images.jl (which is built on top of ImageCore), the color dimension always has to be placed first

There's a permuteddimsview to lazily-permute the dimensions when that's not true.

The various new views (normedview, rawview, channelview, colorview) will reinterpret a block of memory when it's feasible to do so, and otherwise return a wrapper that gives the same functionality. So if you use those functions, getting out an Array means that you can safely pass it directly to C++ without making copies.

If you don't get out an Array, but get some kind of AbstractArray, beware that you probably can't pass such an array as a Ptr to C++ routines and expect it to work.

@timholy
Thanks for sharing the info here. Is this documented in the Images project Wiki? It would be good to keep such relevant information in a Wiki. Perhaps we can have a new section either on OpenCV.jl or Images.jl that documents how to interact between these and other relevant graphics/image-processing Julia packages.

Or should we document such info in the JuliaImages Wiki?

I think it's pretty thoroughly documented at http://juliaimages.github.io/latest/. That said, if you notice stuff that's missing, by all means file an issue.

I'm not against the idea of a Wiki, but that's not where I've put my efforts thus far.

The docs at JuliaImages look good. I have no time myself for the Wiki, but thought that it might be something kvmanohar22 could help with if useful for the project...

maxruby · 2017-03-07T13:49:13Z

test/jl/tests.jl

+   println("test 6: Conversion of images from Images.jl to OpenCV Mat")
+   filename = joinpath(Pkg.dir("OpenCV"), "./test/images/mandrill.jpg")
+   image = imread(filename)
+   dst = Mat(int(rows(image)), int(cols(image)), CV_8UC1)


I believeint is deprecated in 0.6.0, so we need to use Int(round(rows(image))) instead. Please check this and confirm that test5 and test6 pass on your system.

into cpp_templates

maxruby · 2017-03-07T22:25:06Z

test/jl/tests.jl

+   dst = Mat(Int(round(rows(image))), Int(round(cols(image))), CV_8UC1)
   dst = convertToMat(image)
 end



Have you run test5 and test6 on your system?

No, I was trying to get over this error. I'll test them now.

Test 5 :
Video is streamed but I get this error in the terminal

test 5: Basic video display: set video properties Frames per second: 10.0 :1:1: error: no matching constructor for initialization of 'cv::Mat' ^ /usr/local/include/opencv2/core/mat.inl.hpp:339:6: note: candidate constructor not viable: no known conversion from 'long *' to 'int' for 2nd argument; dereference the argument with * Mat::Mat(int _rows, int _cols, int _type) ^ /usr/local/include/opencv2/core/mat.inl.hpp:373:6: note: candidate constructor not viable: no known conversion from 'long *' to 'const int *' for 2nd argument Mat::Mat(int _dims, const int* _sz, int _type) ^ /usr/local/include/opencv2/core/mat.inl.hpp:364:6: note: candidate constructor not viable: no known conversion from 'long' to 'Size' (aka 'Size_<int>') for 1st argument Mat::Mat(Size _sz, int _type, const Scalar& _s) ^ /usr/local/include/opencv2/core/mat.inl.hpp:398:6: note: candidate constructor not viable: no known conversion from 'long' to 'const std::vector<int>' for 1st argument Mat::Mat(const std::vector<int>& _sz, int _type, const Scalar& _s) ^ /usr/local/include/opencv2/core/mat.hpp:894:5: note: candidate constructor not viable: no known conversion from 'long' to 'const std::vector<int>' for 1st argument Mat(const std::vector<int>& sizes, int type, void* data, const size_t* steps=0); ^ /usr/local/include/opencv2/core/mat.hpp:906:5: note: candidate constructor not viable: no known conversion from 'long' to 'const cv::Mat' for 1st argument Mat(const Mat& m, const Range& rowRange, const Range& colRange=Range::all()); ^ /usr/local/include/opencv2/core/mat.inl.hpp:459:6: note: candidate constructor not viable: no known conversion from 'long' to 'Size' (aka 'Size_<int>') for 1st argument Mat::Mat(Size _sz, int _type, void* _data, size_t _step) ^ /usr/local/include/opencv2/core/mat.inl.hpp:492:6: note: candidate constructor template not viable: requires at most 2 arguments, but 3 were provided Mat::Mat(const std::vector<_Tp>& vec, bool copyData) ^ /usr/local/include/opencv2/core/mat.inl.hpp:509:6: note: candidate constructor template not viable: requires at most 2 arguments, but 3 were provided Mat::Mat(const Vec<_Tp, n>& vec, bool copyData) ^ /usr/local/include/opencv2/core/mat.inl.hpp:525:6: note: candidate constructor template not viable: requires at most 2 arguments, but 3 were provided Mat::Mat(const Matx<_Tp,m,n>& M, bool copyData) ^ /usr/local/include/opencv2/core/mat.inl.hpp:541:6: note: candidate constructor template not viable: requires at most 2 arguments, but 3 were provided Mat::Mat(const Point_<_Tp>& pt, bool copyData) ^ /usr/local/include/opencv2/core/mat.inl.hpp:560:6: note: candidate constructor template not viable: requires at most 2 arguments, but 3 were provided Mat::Mat(const Point3_<_Tp>& pt, bool copyData) ^ /usr/local/include/opencv2/core/mat.inl.hpp:580:6: note: candidate constructor template not viable: requires single argument 'commaInitializer', but 3 arguments were provided Mat::Mat(const MatCommaInitializer_<_Tp>& commaInitializer) ^ /usr/local/include/opencv2/core/mat.inl.hpp:390:6: note: candidate constructor not viable: requires 2 arguments, but 3 were provided Mat::Mat(const std::vector<int>& _sz, int _type) ^ /usr/local/include/opencv2/core/mat.hpp:936:5: note: candidate constructor not viable: requires 2 arguments, but 3 were provided Mat(const Mat& m, const std::vector<Range>& ranges); ^ /usr/local/include/opencv2/core/mat.hpp:926:5: note: candidate constructor not viable: requires 2 arguments, but 3 were provided Mat(const Mat& m, const Range* ranges); ^ /usr/local/include/opencv2/core/mat.hpp:916:5: note: candidate constructor not viable: requires 2 arguments, but 3 were provided Mat(const Mat& m, const Rect& roi); ^ /usr/local/include/opencv2/core/mat.inl.hpp:356:6: note: candidate constructor not viable: requires 2 arguments, but 3 were provided Mat::Mat(Size _sz, int _type) ^ /usr/local/include/opencv2/core/mat.inl.hpp:381:6: note: candidate constructor not viable: requires 4 arguments, but 3 were provided Mat::Mat(int _dims, const int* _sz, int _type, const Scalar& _s) ^ /usr/local/include/opencv2/core/mat.inl.hpp:347:6: note: candidate constructor not viable: requires 4 arguments, but 3 were provided Mat::Mat(int _rows, int _cols, int _type, const Scalar& _s) ^ /usr/local/include/opencv2/core/mat.inl.hpp:407:6: note: candidate constructor not viable: requires single argument 'm', but 3 arguments were provided Mat::Mat(const Mat& m) ^ /usr/local/include/opencv2/core/cuda.inl.hpp:621:6: note: candidate constructor not viable: requires single argument 'm', but 3 arguments were provided Mat::Mat(const cuda::GpuMat& m) ^ /usr/local/include/opencv2/core/mat.inl.hpp:1179:6: note: candidate constructor not viable: requires single argument 'm', but 3 arguments were provided Mat::Mat(Mat&& m) ^ /usr/local/include/opencv2/core/mat.hpp:880:5: note: candidate constructor not viable: requires at least 4 arguments, but 3 were provided Mat(int ndims, const int* sizes, int type, void* data, const size_t* steps=0); ^ /usr/local/include/opencv2/core/mat.inl.hpp:426:6: note: candidate constructor not viable: requires at least 4 arguments, but 3 were provided Mat::Mat(int _rows, int _cols, int _type, void* _data, size_t _step) ^ /usr/local/include/opencv2/core/mat.inl.hpp:333:6: note: candidate constructor not viable: requires 0 arguments, but 3 were provided Mat::Mat()

and I was unable to test Test 6, there seems to be a bug with Images.jl which is used in convertToMat function for test 6. I have filed an issue here.

I know what is going on. You need to provide at least one input with a T declaration from which the C++ compiler can use to infer the type. For example:

If you declare the functions like this:

template <typename T1> inline T1 ptr_val(cv::Mat &img, int row, int col) { return(static_cast<T1>(img.at<T1>(row, col))); } template <typename T2> inline void ptr_val3(cv::Mat &img, int row, int col, std::vector<T2> vec) { for(int i = 0; i < 3; ++i) { img.at<cv::Vec<T2, 3>>(row, col)[i] = static_cast<T2>(vec[i]); } } // This works julia> img = Mat(500,500, CV_8UC3) (class cv::Mat) julia> @cxx ptr_val(img, 10, 10, 255) julia> @cxx ptr_val(img, 10, 10, 233.4) julia> @cxx ptr_val(img, 10, 10, 233.4344343) // This also works julia> vec1 = tostdvec([1,2]) (class std::__1::vector<long, class std::__1::allocator<long> >) julia> @cxx ptr_val3(img, 10, 10, vec1) julia> vec1 = tostdvec([1,2])

Without these changes, I get as expected errors like this:

@cxx ptr_val(img, 10, 10) :1:1: error: no matching function for call to 'ptr_val' ^ __cxxjl_87.cpp:3:11: note: candidate template ignored: couldn't infer template argument 'T1' inline T1 ptr_val(cv::Mat &img, int row, int col) ^ __cxxjl_87.cpp:9:13: note: candidate function template not viable: requires 4 arguments, but 3 were provided inline void ptr_val(cv::Mat &img, int row, int col, T2 val)

However, this is surely not the goal - just an illustration of type inference.
The goal here is that the compiler infers the T type based on the Mat type, in other words, we need to use cv::Mat<_Tp> as illustrated in the OpenCV docs.

So, basically we need to modify src/OpenCV_core.jl to include the wrappers for templates ? Are you working on this ? If not I would like to dive in!

By all means dive in and set up the wrappers!

I did have a quick look yesterday evening at implementing cv::Mat<_Tp> but got errors probably because it was not as straightforward as it first seems.

I do not actually have much time right now to work on this project (definitely not during the week) - but I would like to make sure OpenCV.jl continues to work properly and aim always to enhance it in areas where Julia offers unique advantages.

maxruby · 2017-03-09T19:48:43Z

test/runtests.jl

@@ -1,4 +1,5 @@
 using OpenCV
 using Base.Test
+using Images


It would be preferable that you first test the new generic Mat constructors with OpenCV.jl native methods. Once completed you can submit a separate PR with support for Images and any other relevant packages.

Currently I'm working on adding the wrappers for template constructors in src/OpenCV_core.jl for Mat. I'll open a new PR for this. Once this is done and merged only then will I be able to continue with the test 6 using support of other packages or native methods. Images dependency was used because convertToMat method uses Images.

kvmanohar22 · 2017-03-10T07:17:56Z

I came up with a different approach to deal with this error :

@cxx ptr_val(img, 10, 10)
:1:1: error: no matching function for call to 'ptr_val'
^
__cxxjl_87.cpp:3:11: note: candidate template ignored: couldn't infer template argument 'T1'
inline T1 ptr_val(cv::Mat &img, int row, int col)
^

something like this in src/OpenCV_Mat.jl :

  # Grayscale and binary images
  if (cvtypeval(img) == CV_8UC1)
    @cxx ptr_val<unsigned char>(img, row, col, value);
  elseif (cvtypeval(img) == CV_16SC1)
    @cxx ptr_val<short>(img, row, col, value);
  elseif (cvtypeval(img) == CV_32SC1)
    @cxx ptr_val<unsigned short>(img, row, col, value);
  elseif (cvtypeval(img) == CV_32FC1)
    @cxx ptr_val<float>(img, row, col, value);
  elseif (cvtypeval(img) == CV_64FC1)
    @cxx ptr_val<double>(img, row, col, value);

but this seems not to be supported yet in Cxx .
I get the following error for a sample function :

julia> cxx""" #include <iostream>
       template <typename T>
       inline void _sum(int const& a, int const& b){
           T val = 0;
           val += static_cast<T>(a);
           val += static_cast<T>(b);
           
           std::cout<<val<<std::endl;
       }
       """
true

julia> @cxx _sum<int>(2,3);
ERROR: Unrecognized CPP Expression _sum < int > (2, 3) (comparison)

Simultaneously I'm working on implementing cv::Mat_<_Tp>. I think both these implementations accomplish the same. Any suggestions how to take this forward ?

maxruby · 2017-03-10T08:15:28Z

Did you try the following?

cxx""" #include <iostream>
       template <typename T>
       inline void _sum(T const& a, T const& b){
           T val = static_cast<T>(0);
           val += static_cast<T>(a);
           val += static_cast<T>(b);         
           std::cout<<val<<std::endl;
       }
"""

and this one might be tricky because I am not sure how Cxxhandles img.at<T>

template <typename T>
inline void ptr_val(cv::Mat &img, int row, int col, T val)
{		  
    T value = static_cast<T>(val);
    img.at<T>(row, col) = value;
}

kvmanohar22 · 2017-03-10T08:36:48Z

Did you try the following?

Yep, that works. The only problem is when we don't specify the type in the arguements list.

In the case of extending functionality of Cxx to support,

 julia> @cxx _sum<int>(2,3);

we can address the present issue with the above type of implementation without the need of implementing Mat_<_Tp>.

maxruby · 2017-03-10T10:51:15Z

Yep, that works. The only problem is when we don't specify the type in the arguements list.

Good to know. But the question is why do we need to specify in advance that you are passing an intor double? I thought the whole point is not to worry which type you pass to the function...

kvmanohar22 · 2017-03-10T11:05:49Z

Yes, with the implementation of Mat_<_Tp>, we wouldn't need to specify the type. I'm currently working on this.

kvmanohar22 · 2017-03-17T01:44:24Z

I'm coming across an unusual error :

function convertToMat(image)
    img = permuteddimsview(channelview(image), (2,3,1))
    cd = Base.size(channelview(image))[1] > 3 ? 1 : 3
    _rows = Base.size(image, 1)
    _cols = Base.size(image, 2)

    if (typeof(img[1,1,1].i) == UInt8)
       if (cd == 3) 
         mat = Mat(_rows, _cols, CV_8UC3);
       end
    end
end

I get the following error which is something unexpected.

:1:1: error: no matching constructor for initialization of 'cv::Mat'
^
/usr/local/include/opencv2/core/mat.inl.hpp:339:6: note: candidate constructor not viable: no known conversion from 'long *' to 'int' for 2nd argument; dereference the argument with *
Mat::Mat(int _rows, int _cols, int _type)
     ^

I could boil down the error to this,
Base.size(image, 1) and Base.size(image, 2) is causing the error.
This was evident when I replaced _rows and _cols with 512, 512 in the Mat() constructor.

maxruby · 2017-03-17T07:52:59Z

I don't have access now to my julia/openCV.jl setup, but the error clearly says that long * cannot be used instead of int for the 2nd argument. To me it seems to suggest that Base.size is returning an Int64 instead of an expected Int32for the covertToMat function to work. I am not sure if this will fix the error, but can you test mat = Mat(convert(Int32, _rows), convert(Int32, _cols), CV_8UC3);?

kvmanohar22 · 2017-03-17T08:06:58Z

mat = Mat(convert(Int32, _rows), convert(Int32, _cols), CV_8UC3);

I had tried that earlier, but there was no luck with that.

maxruby · 2017-03-17T08:14:28Z

OK. Let me try to understand better the problem. Can you check what type exactly does Base.size(image, 1) return in the julia REPL? And if you convert to Int32 using convert, what does it show?

kvmanohar22 · 2017-03-17T08:20:04Z

what type exactly does Base.size(image, 1) return

Int64

convert to Int32 using convert

test 6: Conversion of images from Images.jl to OpenCV Mat
ERROR: LoadError: LoadError: MethodError: no method matching Mat(::Int32, ::Int32, ::Int64)
Closest candidates are:
  Mat(::Any, !Matched::Int64, ::Any) at /home/kv/.julia/v0.6/OpenCV/./src/OpenCV_core.jl:368
  Mat(::Any, ::Any) at /home/kv/.julia/v0.6/OpenCV/./src/OpenCV_core.jl:385
  Mat(!Matched::Int64, !Matched::Ptr{Int64}, ::Int64) at /home/kv/.julia/v0.6/OpenCV/./src/OpenCV_core.jl:375
  ...
Stacktrace:
 [1] convertToMat(::Array{ColorTypes.RGB{FixedPointNumbers.Normed{UInt8,8}},2}) at /home/kv/.julia/v0.6/OpenCV/./src/OpenCV_ImagesSupport.jl:17
 [2] test6() at /home/kv/.julia/v0.6/OpenCV/./test/jl/tests.jl:146
 [3] include_from_node1(::String) at ./loading.jl:539
 [4] include(::String) at ./sysimg.jl:14
 [5] include_from_node1(::String) at ./loading.jl:539
 [6] include(::String) at ./sysimg.jl:14
 [7] process_options(::Base.JLOptions) at ./client.jl:305
 [8] _start() at ./client.jl:371
while loading /home/kv/.julia/v0.6/OpenCV/./test/jl/tests.jl, in expression starting on line 149
while loading /home/kv/gsoc/OpenCV.jl/test/runtests.jl, in expression starting on line 6

timholy · 2017-03-17T08:40:52Z

That's true on a 64-bit CPU; it's Int32 on a 32-bit CPU. Safest is to use Int.

kvmanohar22 · 2017-03-17T08:47:52Z

but I don't get any errors while doing
mat = Mat(512, 512, CV_8UC3)
the problem is only when I deduce the size using Base.size() and use it in the function :
mat = Mat(_rows, _cols, CV_8UC3).
I'm unable to figure out why ?

maxruby · 2017-03-17T09:15:34Z

That's true on a 64-bit CPU; it's Int32 on a 32-bit CPU. Safest is to use Int.

What happens when you provide convert(Int, Base.size(img, 1)) and convert(Int, Base.size(img, 2))?

kvmanohar22 · 2017-03-17T11:39:43Z

What happens when you provide convert(Int, Base.size(img, 1)) and convert(Int, Base.size(img, 2))?

I get the following error :

:1:1: error: no matching constructor for initialization of 'cv::Mat'
^
/usr/local/include/opencv2/core/mat.inl.hpp:339:6: note: candidate constructor not viable: no known conversion from 'long *' to 'int' for 2nd argument; dereference the argument with *
Mat::Mat(int _rows, int _cols, int _type)
     ^

maxruby · 2017-03-17T22:33:07Z

See the following posts (old but still relevant)

maxruby · 2017-03-18T10:43:46Z

src/OpenCV_Mat.jl

-template <typename T2>
-inline void ptr_val(cv::Mat<T2> &img, int row, int col, double val)
+template <typename T>
+inline void ptr_val(cv::Mat &img, int row, int col, double val)


I have had a look at the possible implementation of cv::Mat templates as well, but as discussed before the implementation here does not work. When executed in the REPL, I get couldn't infer template argument T:

julia> @cxx ptr_val(img, 10,10, 200.5) :1:1: error: no matching function for call to 'ptr_val' __cxxjl_93.cpp:2:13: note: candidate template ignored: couldn't infer template argument 'T' inline void ptr_val(cv::Mat &img, int row, int col, double val) {

I will add some comments on my observations and suggestion for an implementation using a combination of std::vector and _InputArray. OpenCVby design (due to performance reasons) uses non-templated Mat constructors.

Yes, that would be great. I'm stuck with the implementation of Mat_<_Tp> constructers.

maxruby · 2017-03-19T16:43:08Z

I found a solution for conversion of image arrays from JuliaImages to Mat using templates and have written wrappers to test it. The key is to use cv::DataType<T>::type when constructing cv::Mat because otherwise you always need to specify the image type, e.g., CV_8UC1.
I will first commit this tonight, then we can look into the next steps.

using c++ templates

e869082

Adding tests for convertToMat method

3501983

maxruby requested changes Mar 5, 2017

View reviewed changes

maxruby reviewed Mar 5, 2017

View reviewed changes

fix bug

3f94074

maxruby reviewed Mar 6, 2017

View reviewed changes

kvmanohar22 and others added 2 commits March 7, 2017 12:38

use native functions

45927e6

Merge branch 'master' into cpp_templates

85ffe6c

maxruby reviewed Mar 7, 2017

View reviewed changes

kvmanohar22 added 2 commits March 8, 2017 03:16

deprecated int issue in 0.6.0

7d4b6aa

Merge branch 'cpp_templates' of https://github.com/kvmanohar22/OpenCV.jl

6310f3f

into cpp_templates

maxruby reviewed Mar 7, 2017

View reviewed changes

maxruby assigned kvmanohar22 Mar 8, 2017

maxruby added the enhancement label Mar 8, 2017

kvmanohar22 added 2 commits March 9, 2017 02:48

update templates

3458803

Adding Images dependency for tests

d94f8ad

maxruby reviewed Mar 9, 2017

View reviewed changes

new approach to templates

c4729c6

remove deprecated functionalities

42447f7

maxruby reviewed Mar 18, 2017

View reviewed changes

[WIP] Using c++ templates #18

Are you sure you want to change the base?

[WIP] Using c++ templates #18

Uh oh!

Conversation

kvmanohar22 commented Mar 3, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

maxruby commented Mar 4, 2017

Uh oh!

kvmanohar22 commented Mar 4, 2017

Uh oh!

maxruby commented Mar 4, 2017

Uh oh!

maxruby commented Mar 5, 2017

Uh oh!

maxruby left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

maxruby Mar 7, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

kvmanohar22 Mar 8, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

kvmanohar22 Mar 10, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

kvmanohar22 commented Mar 10, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

maxruby commented Mar 10, 2017

Uh oh!

kvmanohar22 commented Mar 10, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

maxruby commented Mar 10, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

kvmanohar22 commented Mar 10, 2017

Uh oh!

kvmanohar22 commented Mar 17, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

maxruby commented Mar 17, 2017

Uh oh!

kvmanohar22 commented Mar 17, 2017

Uh oh!

maxruby commented Mar 17, 2017

Uh oh!

kvmanohar22 commented Mar 17, 2017

kvmanohar22 commented Mar 3, 2017 •

edited

Loading

maxruby Mar 7, 2017 •

edited

Loading

kvmanohar22 Mar 8, 2017 •

edited

Loading

kvmanohar22 Mar 10, 2017 •

edited

Loading

kvmanohar22 commented Mar 10, 2017 •

edited

Loading

kvmanohar22 commented Mar 10, 2017 •

edited

Loading

maxruby commented Mar 10, 2017 •

edited

Loading

kvmanohar22 commented Mar 17, 2017 •

edited

Loading

maxruby commented Mar 17, 2017 •

edited

Loading

maxruby Mar 18, 2017 •

edited

Loading