sonots

Usually, located at /usr/local/cuda/bin

Non-Visual Profiler

$ nvprof python train_mnist.py

I prefer to use --print-gpu-trace.

HarshaVardhanBabu

Requirements

1. Install Pylint from Install. If you have anaconda already installed use pip install -U pylint to update the Pylint so that pyreverse is added to the scripts folder.

2. You need to install Graphviz as the pyreverse generates the UML diagrams in dot format and needs the dot.exe provided by Graphviz. Once Graphviz is installed make sure the bin folder is added to the PATH variable so that pyreverse can find it at run time. "the command pyreverse generates the diagrams in all formats that graphviz/dot knows." (Reference

3. Now add the path of python modules for which you want to generate the documentation to PYTHONPATH.

4. Use pyreverse -S <modulename> to generate dot files in the current folder

   Usage:

dentechy

How to automatically start ssh server on boot on Windows Subsystem for Linux

Microsoft partnered with Canonical to create Bash on Ubuntu on Windows, running through a technology called the Windows Subsystem for Linux. Below are instructions on how to set up the ssh server to run automatically at boot.

1. Edit the /etc/ssh/sshd_config file by running the command sudo vi /etc/ssh/sshd_config and do the following
   1. Change Port to 2222 (or any other port above 1000)
   2. Change PasswordAuthentication to yes. This can be changed back to no if ssh keys are setup.
2. Restart the ssh server:
   • sudo service ssh --full-restart
3. With this setup, the ssh server must be turned on every time you run Bash on Ubuntu on Windows, as by default it is off. Use this command to turn it on:

erogol

import torch as th                                                                 
                                                                                   
                                                                                   
class NLL_OHEM(th.nn.NLLLoss):                                                     
    """ Online hard example mining. 
    Needs input from nn.LogSotmax() """                                             
                                                                                   
    def __init__(self, ratio):      
        super(NLL_OHEM, self).__init__(None, True)                                 
        self.ratio = ratio                                                         

jacobkimmel

def make_one_hot(labels, C=2):
    '''
    Converts an integer label torch.autograd.Variable to a one-hot Variable.
    
    Parameters
    ----------
    labels : torch.autograd.Variable of torch.cuda.LongTensor
        N x 1 x H x W, where N is batch size. 
        Each value is an integer representing correct classification.
    C : integer. 

endolith

"""
Train a neural network to implement the discrete Fourier transform
"""
import matplotlib.pyplot as plt
import numpy as np
from tensorflow.keras.layers import Dense
from tensorflow.keras.models import Sequential

N = 32
batch = 10000

bitsgalore

ImageMagick / OpenJPEG build and install notes

I always end up getting this wrong; steps below worked for Linux Mint 19.3 (based on Ubuntu 18.04). Build/installation order is important; JPEG 2000 support in ImageMagick only works if OpenJPEG is found at build time, so we have to start with that. Note that for OpenJPEG an 'openjpeg-dev' Debian package exists. As I'm not entirely sure this is the most up-to-date version, and JPEG 2000 support is important for me, I'm compiling this library from the sources here. Otherwise everything under the 'OpenJPEG' could probably be subsituted by the one-liner sudo at-get install openjpeg-dev).

Cmake