Can you give advice on feature generation/kernel selection?
Picking the right kernel all comes down to understanding your data, and obviously this is highly dependent on your problem.
Hi, i'm using Kali, followed all the steps: Install libboost, compile dlib. If I install libboost-all-dev, dlib can't import libboostpython.so.1.64.0 (as I believe it installs the 1.62 version). In this video we will see how to install the Dlib library for Python 3 on Windows.In this specific tutorial we are going to install Dlib 19.8 for Python 3.6. I don’t know why but cmake built for python fine by skipping “workon cv” and after pointing the “-D PYTHONEXECUTABLE=” line to /usr/bin/python3. I can use virtualenv now after creating the link with ln -s, except my cv2 file went to dist-packages instead of site-packages.
One thing that's sometimes useful is to plot each feature against the target value. You can get an idea of what your overall feature space looks like and maybe tell if a linear kernel is the right solution. But this still hides important information from you. For example, imagine you have two diagonal lines which are very close together and are both the same length. Suppose one line is of the +1 class and the other is the -1 class. Each feature (the x or y coordinate values) by itself tells you almost nothing about which class a point belongs to but together they tell you everything you need to know.
First you need Visual Studio. So download Visual Studio on your Computer. Then install Cmake. Whether you download it from website or use pip: pip install cmake Now install dlib: pip install dlib It may take a while to build wheel for dlib, after a while, it will be successfully installed. Cd dlib mkdir build; cd build; cmake.; cmake — build. Build and install python essentials. Python3 setup.py install. After all this run these to verify. Python3 import dlib.
On the other hand, if you know something about the data you are working with then you can also try and generate your own features. So for example, if your data is a bunch of images and you know that one of your classes contains a lot of lines then you can make a feature that attempts to measure the number of lines in an image using a hough transform or sobel edge filter or whatever. Generally, try and think up features which should be highly correlated with your target value. A good way to do this is to try and actually hand code N solutions to the problem using whatever you know about your data or domain. If you do a good job then you will have N really great features and a linear or rbf kernel will probably do very well when using them.
Python Setup Cmake
Or you can just try a whole bunch of kernels, kernel parameters, and training algorithm options while using cross validation. I.e. when in doubt, use brute force :) There is an example of that kind of thing in the model selection example program.
I am trying to install dlib for which I have already installed cmake.
cmake version 3.10.2
Cmake Dlib Python Download
CMake suite maintained and supported by Kitware (kitware.com/cmake).
After issuing command : pip3 install dlib
It keeps on building wheel for dlib. I waited to ~1hr but it didnt installed and was just showing building wheel for dlib. The nano however got freezed and wasnt responding so I restarted it but dlib was not installed.
How to properly installed dlib. Does the installtion takes this much time.?
Cmake Python Example
Python Cmake Module