davegreenwood

# %%

import torch
import torch.nn.functional as F

# compute rotation vector to get from a to b
def angleBetweenRT(a, b):

    # find rotation axis
    rotvec = torch.cross(a, b)

davegreenwood

#!/bin/bash

# Make executable with chmod +x <<filename.sh>>
# based on:
# https://www.freecodecamp.org/news/automate-project-github-setup-mac/
#
# but using a github personal access token:
# https://help.github.com/en/github/authenticating-to-github/creating-a-personal-access-token

davegreenwood

"""Fit a 2d line to data using auto differentiation to learn w, b."""
# %%
import torch
import matplotlib.pyplot as plt

# size of data - more data wil give a more accurate result
N = 500

# we can add noise to data to make it more realistic
noise = torch.randn(N)

davegreenwood

CMPGPUSVR hints and tips

Server is run on a be polite policy - it's not a right but a privilege :)

Etiquette

There are two main considerations for users.

GPU

davegreenwood

*PPD-Adobe: "4.3"
*%
*%   PPD file for Dell 1320 netork color printer
*%
*%   Based on orginal file:
*%
*%   FX DocuPrint C525 A-AP PPD file for CUPS.
*%
*%   Copyright 2005 by Fuji Xerox Co.,Ltd.
*%

davegreenwood

from plotly.offline import init_notebook_mode, iplot
import plotly.graph_objs as go
init_notebook_mode()

# x, y is point data, image is a PIL image.

trace0 = go.Scatter(
    x=x, y=y, mode='markers',
    hovertext=[str(i) for i in range(len(x))],
    hoverinfo="text",)

davegreenwood

How to get the proper OpenGL projection matrix from the OpenCV camera calibration values?

For the simple common case where the OpenCV camera matrix has the form:

|fx  0 cx|
|0  fy cy|
|0  0   1|

The corresponding OpenGL projection matrix can be computed like this:

davegreenwood

{
	// Place your snippets for python here. Each snippet is defined under a snippet name and has a prefix, body and
	// description. The prefix is what is used to trigger the snippet and the body will be expanded and inserted. Possible variables are:
	// $1, $2 for tab stops, $0 for the final cursor position, and ${1:label}, ${2:another} for placeholders. Placeholders with the
	// same ids are connected.
	// Example:
	// "Print to console": {
	// 	"prefix": "log",
	// 	"body": [
	// 		"console.log('$1');",

davegreenwood

Setting up jupyter notebook

First install jupyter themes to change appearance.

pip install jupyterthemes

Then, set the theme to taste

jt -t onedork \

davegreenwood

# %% reorder the vertices
import numpy as np

# %%
with open("oldorder.obj") as fid:
    lines = fid.readlines()
print(lines)

#  %%