Related
Its possible to print a full array in the console with:
import sys
import numpy
numpy.set_printoptions(threshold=sys.maxsize)
but is there also an option to export a kind of "number chart image"? e.g.
import numpy as np
numberchart = np.range(100)
WANTED_RESULT.png
I plotted with matplotlib some kind of a heatmap, but Iam looking for image format like .png
harvest = np.array([[0.8, 2.4, 2.5, 3.9, 0.0, 4.0, 0.0],
[2.4, 0.0, 4.0, 1.0, 2.7, 0.0, 0.0],
[1.1, 2.4, 0.8, 4.3, 1.9, 4.4, 0.0],
[0.6, 0.0, 0.3, 0.0, 3.1, 0.0, 0.0],
[0.7, 1.7, 0.6, 2.6, 2.2, 6.2, 0.0],
[1.3, 1.2, 0.0, 0.0, 0.0, 3.2, 5.1],
[0.1, 2.0, 0.0, 1.4, 0.0, 1.9, 6.3]])
fig, ax = plt.subplots()
im = ax.imshow(harvest)
y, x = harvest.shape
ax.set_xticks(np.arange(x))
ax.set_yticks(np.arange(y))
plt.setp(ax.get_xticklabels())#, rotation=45, ha="right",
#rotation_mode="anchor")
# Loop over data dimensions and create text annotations.
for i in range(x):
for j in range(y):
text = ax.text(j, i, harvest[i, j],
ha="center", va="center", color="w")
ax.set_title("NumberChart")
fig.tight_layout()
plt.show()
I did some minor changes to your code. It is more like a work around but does what you hope I beleive:
import matplotlib.pyplot as plt
harvest = np.array([[0.8, 2.4, 2.5, 3.9, 0.0, 4.0, 0.0],
[2.4, 0.0, 4.0, 1.0, 2.7, 0.0, 0.0],
[1.1, 2.4, 0.8, 4.3, 1.9, 4.4, 0.0],
[0.6, 0.0, 0.3, 0.0, 3.1, 0.0, 0.0],
[0.7, 1.7, 0.6, 2.6, 2.2, 6.2, 0.0],
[1.3, 1.2, 0.0, 0.0, 0.0, 3.2, 5.1],
[0.1, 2.0, 0.0, 1.4, 0.0, 1.9, 6.3]])
fig, ax = plt.subplots()
im = ax.imshow(harvest*0, cmap="Greys")
y, x = harvest.shape
ax.set_xticks(np.arange(x))
ax.set_yticks(np.arange(y))
plt.setp(ax.get_xticklabels())#, rotation=45, ha="right",
#rotation_mode="anchor")
# Loop over data dimensions and create text annotations.
for i in range(x):
for j in range(y):
text = ax.text(j, i, harvest[i, j],
ha="center", va="center", color="k")
ax.set_xticks(np.arange(-.5, harvest.shape[0], 1), minor=True)
ax.set_yticks(np.arange(-.5, harvest.shape[1], 1), minor=True)
plt.tick_params(
axis='x',
which='both',
bottom=False,
top=False,
labelbottom=False)
plt.tick_params(
axis='y',
which='both',
left=False,
top=False,
labelleft=False)
ax.grid(which='minor', color='k', linestyle='-', linewidth=1)
ax.set_title("NumberChart")
fig.tight_layout()
plt.show()
I work with webgl and modify the shaders (vs.glsls and fs.glsl) to understand the GLSL and graphic programming. I have a model that I want to scale, rotate and translate. Scaling and rotating works fine but when I multiply the translation matrix, the result is weird. I know this is a very basic question, but I am missing something and I need to find it out.
my model is infinitely stretchered through the y axis.
The white area is supposed to be the eye of the model:
this is my vertex shader code:
mat4 rX = mat4 (
1.0, 0.0, 0.0, 0.0,
0.0, 0.0, -1.0, 0.0,
0.0, 1.0, 0.0, 0.0,
0.0, 0.0, 0.0, 1.0
);
mat4 rZ = mat4 (
0.0, 1.0, 0.0, 0.0,
-1.0, 0.0, 0.0, 0.0,
0.0, 0.0, 1.0, 0.0,
0.0, 0.0, 0.0, 1.0
);
mat4 eyeScale = mat4 (
.50,0.0,0.0,0.0,
0.0,.50,0.0,0.0,
0.0,0.0,.50,0.0,
0.0,0.0,0.0,1.0
);
mat4 eyeTrans = mat4(
1.0,0.0,0.0,0.0,
0.0,1.0,0.0,4.0,
0.0,0.0,1.0,0.0,
0.0,0.0,0.0,1.0
);
mat4 iR = eyeTrans*rZ*rX*eyeScale;
gl_Position = projectionMatrix * modelViewMatrix *iR* vec4(position, 1.0);
}
You swapped rows and columns, when you set up the translation matrix
Change it to:
mat4 eyeTrans = mat4(
1.0, 0.0, 0.0, 0.0,
0.0, 1.0, 0.0, 0.0,
0.0, 0.0, 1.0, 0.0,
0.0, 4.0, 0.0, 1.0
);
A 4*4 matrix looks like this:
c0 c1 c2 c3 c0 c1 c2 c3
[ Xx Yx Zx Tx ] [ 0 4 8 12 ]
[ Xy Yy Zy Ty ] [ 1 5 9 13 ]
[ Xz Yz Zz Tz ] [ 2 6 10 14 ]
[ 0 0 0 1 ] [ 3 7 11 15 ]
In GLSL the columns are addressed like this:
vec4 c0 = eyeTrans[0].xyzw;
vec4 c1 = eyeTrans[1].xyzw;
vec4 c2 = eyeTrans[2].xyzw;
vec4 c3 = eyeTrans[3].xyzw;
And the memory image of a 4*4 matrix looks like this:
[ Xx, Xy, Xz, 0, Yx, Yy, Yz, 0, Zx, Zy, Zz, 0, Tx, Ty, Tz, 1 ]
See further:
GLSL 4×4 Matrix Fields
I am trying to create billboard using three.js. I tried using THREE.Sprite however size of sprite changes with distance as I am using perspective projection. I tried to create a plane using custom shaderMaterial with custom shader.
Shader code for scaling billboard
vec4 gl_Position = projectionMatrix * (modelViewMatrix * vec4(0.0, 0.0, 0.0, 1.0) + vec4(position.x, position.y, 0.0, 0.0));'
same for non scaling billboard, which I am trying implement
float distance = length(modelViewMatrix * vec4(0.0, 0.0, 0.0, 1.0));
vec4 newPosition = projectionMatrix * (modelViewMatrix * vec4(0.0, 0.0, 0.0, 1.0) + vec4(position.x, position.y, 0.0, 0.0));
gl_Position= newPosition*distance;
Unfortunately there is no changes in the output. Its still scaled billboard. Any suggestion that might fix the problem?
Is it possible to write a 5x5 kernel to process the limited color range into the full range?
This is my sample bitonal kernel, and I don't know what values to use and where to achieve this color expansion:
Grayscale
{ 0.3, 0.3, 0.3, 0.0, 0.0 }
{ 0.6, 0.6, 0.6, 0.0, 0.0 }
{ 0.1, 0.1, 0.1, 0.0, 0.0 }
{ 0.0, 0.0, 0.0, 1.0, 0.0 }
{ 0.0, 0.0, 0.0, 0.0, 1.0 }
I would like RGB color expansion RGB 16-235 => 0-255
However i need the kernel matrix because I am not processing the image but I'm passing the matrix to a windows API function (undocumented: SetMagnificationDesktopColorEffect).
I cannot do a simple subtract/divide/multiply on the pixels. I do not have them.
You can basically do it without kernel by substracting 16 from your image and then dividing it by 219. Then you will have normalized to 1 image which you have to multiply by 255 to get 255 intensity range representation.
So I'm writing some WebGL, no THREE.JS. I'm trying to render a cube, with a single texture mapped to every face of the cube. In my code where I set up my attributes I have something like:
var vertices = new Float32Array([
// x, y, z u, v
1.0, 1.0, 1.0, /* v0 right top front */ 1.0, 1.0,
-1.0, 1.0, 1.0, /* v1 left top front */ 0.0, 1.0,
-1.0, -1.0, 1.0, /* v2 left bottom front */ 0.0, 0.0,
1.0, -1.0, 1.0, /* v3 right bottom front */ 1.0, 0.0,
// u's switch for back faces
1.0, -1.0, -1.0, /* v4 right bottom back */ 0.0, 0.0,
1.0, 1.0, -1.0, /* v5 right top back */ 0.0, 1.0,
-1.0, 1.0, -1.0, /* v6 left top back */ 1.0, 1.0,
-1.0, -1.0, -1.0, /* v7 left bottom back */ 1.0, 0.0
]);
// the pairs of vertex triples
// 3 vertices = 1 triangle
// 2 triangles = 1 quad = 1 face
var indices = new Uint8Array([
0, 1, 2, 0, 2, 3, // front
0, 3, 4, 0, 4, 5, // right
//0, 5, 6, 0, 6, 1, // top
1, 6, 7, 1, 7, 2, // left
//7, 4, 3, 7, 3, 2, // bottom
4, 7, 6, 4, 6, 5 // back
]);
I wind up with a cube with the texture reflected for the right and left faces, which is fine. For the top and the bottom, I have no faces because of the two commented out lines. When I comment them in, the faces don't have the texture sampled as I expected. Sure enough, if you look at the indices for the top face for instance, and the UV coordinates that they would have:
index | u | v
0 | 1.0 | 1.0
1 | 0.0 | 1.0
5 | 0.0 | 1.0
6 | 1.0 | 1.0
So we can see that index 1 and 5 (also, 0 and 6) have the same UV coordinates, so of course it wont look right on a quad.
I've been trying to draw out on paper, but I can't change the UV's without messing up another face's coordinates. What I'm wondering is: is it possible to use ELEMENT_ARRAY_BUFFERs to map UV coordinates on a cube, or do I need to use more data and draw using an ARRAY_BUFFER?
== EDIT ==
Looks like a dupe: OpenGL ES - texture map all faces of an 8 vertex cube?
Hate to answer my own question, but based on the hint here, I was able to get it to work by using 24 vertices instead of 8. I can use 24 instead of 36 because I'm repeating indices in my ELEMENT_ARRAY_BUFFER (something I wouldn't be able to do with just an ARRAY_BUFFER).
var vertices = new Float32Array([
// x, y, z, u, v
// front face (z: +1)
1.0, 1.0, 1.0, 1.0, 1.0, // top right
-1.0, 1.0, 1.0, 0.0, 1.0, // top left
-1.0, -1.0, 1.0, 0.0, 0.0, // bottom left
1.0, -1.0, 1.0, 1.0, 0.0, // bottom right
// right face (x: +1)
1.0, 1.0, -1.0, 1.0, 1.0, // top right
1.0, 1.0, 1.0, 0.0, 1.0, // top left
1.0, -1.0, 1.0, 0.0, 0.0, // bottom left
1.0, -1.0, -1.0, 1.0, 0.0, // bottom right
// top face (y: +1)
1.0, 1.0, -1.0, 1.0, 1.0, // top right
-1.0, 1.0, -1.0, 0.0, 1.0, // top left
-1.0, 1.0, 1.0, 0.0, 0.0, // bottom left
1.0, 1.0, 1.0, 1.0, 0.0, // bottom right
// left face (x: -1)
-1.0, 1.0, 1.0, 1.0, 1.0, // top right
-1.0, 1.0, -1.0, 0.0, 1.0, // top left
-1.0, -1.0, -1.0, 0.0, 0.0, // bottom left
-1.0, -1.0, 1.0, 1.0, 0.0, // bottom right
// bottom face (y: -1)
1.0, -1.0, 1.0, 1.0, 1.0, // top right
-1.0, -1.0, 1.0, 0.0, 1.0, // top left
-1.0, -1.0, -1.0, 0.0, 0.0, // bottom left
1.0, -1.0, -1.0, 1.0, 0.0, // bottom right
// back face (x: -1)
-1.0, 1.0, -1.0, 1.0, 1.0, // top right
1.0, 1.0, -1.0, 0.0, 1.0, // top left
1.0, -1.0, -1.0, 0.0, 0.0, // bottom left
-1.0, -1.0, -1.0, 1.0, 0.0 // bottom right
]);
// the pairs of vertex triples
// 3 vertices = 1 triangle
// 2 triangles = 1 quad = 1 face
var indices = new Uint8Array([
0, 1, 2, 0, 2, 3,
4, 5, 6, 4, 6, 7,
8, 9, 10, 8, 10, 11,
12, 13, 14, 12, 14, 15,
16, 17, 18, 16, 18, 19,
20, 21, 22, 20, 22, 23
]);
Scroll above code example ^
The number of vertices can be further reduced, because some indices share the same XYZ and UV coordinates, though if I add normals to my interleaved buffer later (or any other attribute) I may need the repeated values.