Notebook Gallery

Links to the best IPython and Jupyter Notebooks.

CNN (keras example) on mnist (from keras's examples) with convolution visualization

This website does not host notebooks, it only renders notebooks available on other websites.

Markdownテスト

This website does not host notebooks, it only renders notebooks available on other websites.

ジニ係数

Delivered by Fastly , Rendered by Rackspace

Intro to Jupyter!

https://www.digitalocean.com/community/tutorials/how-to-set-up-a-jupyter-notebook-to-run-ipython-on-ubuntu-16-04

Example of How to Allocate Written Question Tagging

The following script shows how work can be fairly allocated to particular individuals based on the number of questions sent to each answering body.

Tokenization

Delivered by Fastly , Rendered by Rackspace

Tournament simulation using PyStan

Main idea: assign score $\alpha_i$ to each team, predict $\sqrt{\Delta}$ where $\Delta$ is the difference in scores using $\sqrt{\Delta} \approx \alpha_i - \alpha_j + h$ where $h$ is the home-field advantage.

ジニ係数

Delivered by Fastly , Rendered by Rackspace

Modulation by CDNN

We use Keras with Tensorflow/Theano engine for testing.

Levanto la data

Es muy interesante ver como algunas provincias son parejas (como Santiago del Estero) en los distintos niveles, mientras otras estan totalmente desbalanceadas (como La Rioja)

CNTK 201B: Hands On Labs Image Recognition

In this hands-on, you will practice the following:

Playing with new 2017 tf.contrib.seq2seq

By this point implementations are quite long, so I put them in model_new.py , while notebook will illustrate the application.

Zero delay the easy way

So-called "zero delay" filters have become popular in synthesizer and audio applications, due to their accurate emulation of analog filter designs and efficient implementation. The techniques used to create zero delay filters are considered standard in the circuit emulation field, ...

The Agate Tutorial

The data we will be using is a copy of the National Registery of Exonerations made on August 28th, 2015. This dataset lists individuals who are known to have been exonerated after having been wrongly convicted in United States courts. ...

Specific simulation techniques

The simplest relation to be exploited is based on the sum of random variables, and it leads to the so-called decompositional simulating approach. Simply put, if $X = Y_1 + \dots + Y_n$ simulating $X$ can be translated into generating ...

Subpackages and functions

where $\alpha$ is the negative log-log slope of power with frequency. $\alpha = 1$ gives a $1/f$ filter.

Demo for gabor filterbank functions

The energy is the sum of squared odd and even filter responses.

Getting Started with RxPY

Rx is about processing streams of events. With Rx you:

Preprocess

When passing in just the image, the MSE doesn't converge as well

Incident handling at Kayako

Our goal is to identify an incident before our customer. It not only helps us in proactively informing our customers on the ongoing issues but also helps us to prepare our Support team better for the incoming queries. It is ...

Support Vector Machines

The following code will train a linear SVM on the dataset and plot the decision boundary learned

It's metaclasses all the way down:

Well, usually you don't.

Random-Walk Logistic Regression in PyMC3

(c) 2017 by Thomas Wiecki -- Quantopian Inc.

Passing spatial data through a KDTree in Spark

My favourite analysis tool is Spark, but it took me a little while to figure out how to work with spatial data in Spark. Google doesn't yield any tutorials on using KDTrees in PySpark, so this is a short tutorial ...

CNTK 204: Sequence to Sequence Networks with Text Data

Andrej Karpathy has a nice visualization of five common paradigms of neural network architectures:

When you write:

NYC Harasment Complaints Data Analysis

First we look at the different description types of complaints in general in 2016:

Biased and Unbiased Point Estimators - Sample mean and variance

where $\mu$ is the mean of the population, defined as:

Comparing Python Clustering Algorithms

To start, lets' lay down some ground rules of what we need a good EDA clustering algorithm to do, then we can set about seeing how the algorithms available stack up.

California electricity capacity analysis

This data analysis provided information for the February 5, 2017, Los Angeles Times story "Californians are paying billions for power they don't need" by documenting California's glut of power and the increasing cost to consumers. It also underpins a complementary ...