Andrew Ng Tutorial: https://web.stanford.edu/class/cs294a/sparseAutoencoder_2011new.pdf UFLDL Exercise: http://ufldl.stanford.edu/wiki/index.php/Exercise:Sparse_Autoencoder A Chinese blog: http://www.cnblogs.com/tornadomeet/archive/2013/03/20/2970724.html Stacked Denoising Autoencoder (paper): http://jmlr.csail.mit.edu/papers/volume11/vincent10a/vincent10a.pdf
Author Archives: czxttkl
Convolutional Neural Network Simple Tutorial
I am going to demonstrate a simple version of convolutional neural network, a type of deep learning structure in this post. Motivation Knowing normal multi-layer neural network (probably the one with input layer, output layer and one hidden layer) is helpful before you proceed reading this post. A good tutorial of MLN can be found …
Continue reading “Convolutional Neural Network Simple Tutorial”
How to run JupyterHub on Ubuntu?
JupyterHub is an Ipython notebook extension which supports multi-user logins. (https://github.com/jupyter/jupyterhub). Installation of JupyterHub, however, requires a bit work due to their incomplete docs and complicated dependencies. Here are the steps I went through to run JupyterHub on my Ubuntu machine: 1. Set alias. JupyerHub will use Python 3.x instead of Python 2.7. But Ubuntu …
How to convert Matlab variables to numpy variables?
Background If you have a Matlab variable, say a matrix, that you want to read in Python. How will you do that? In this post, I am introducing a way that works 100%. 1. Save Matlab variable(s) to ‘.mat’ file. save(‘your_file_name.mat’, ‘first_var’, ‘second_var’,…) 2. In Python, load ‘.mat’ file using `scipy.io`. import scipy.io scipy.io.loadmat(‘your_file_name.mat’) References: …
Continue reading “How to convert Matlab variables to numpy variables?”
Compile Cython Module on Windows
In this post, I am showing how to compile Cython modules on Windows, which often leads to weird error messages. (Cython references: http://cython.org/, http://docs.cython.org/index.html, https://github.com/cython/cython) 1. You should already have correct structure to hold your “.pyx” file. Also, `setup.py` needs to be correctly configured. The following shows an example of `setup.py`: try: from setuptools import …
Life saving tips for PyDev
I wrote a very useful post a year ago, talking about 20 life-saving tips in Eclipse: http://maider.blog.sohu.com/281903297.html. In that context, the tips focus more on efficiency during Java development. In this post, several tips particularly useful for Python developers are collected. I will continue updating the post whenever I find something useful. 1. Press `F2` …
How to type double vertical bar (norm) in latex?
How to type two vertical bars for representing norms? Here is the trick: $\left\Vert something \right\Vert$ The effect will be like this:
How to type gradient descent symbol in latex
Use \nabla to print an upside-down triangle to represent gradient descent. The effect is like this:
How does gradient vanish in Multi-Layer Neural Network?
Background This post reviews how we update weights using the back propagation approach in a neural network. The goal of the review is to illustrate a notorious phenomenon in training MLNN, called “gradient vanish”. Start Let’s suppose that we have a very simple NN structure, with only one unit in each hidden layer, input layer …
Continue reading “How does gradient vanish in Multi-Layer Neural Network?”
How to view all columns of a data.frame in R
If you want to view a large data.frame with a large number of columns in R, you will only see first several columns in the UI. After reading this post, I found that if using utils::View(your_data_frame) , you can view all the columns in a new window. Reference: http://stackoverflow.com/questions/19341853/r-view-does-not-display-all-columns-of-data-frame