amueller

amueller

amueller

{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {

amueller

{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {

amueller

Tutorial Topic

This tutorial aims to provide an introduction to machine learning and scikit-learn "from the ground up". We will start with basic concepts of machine learning and implementing these using scikit-learn. Going in detail through the characteristics of several methods, we will discuss how to pick an algorithm for your application, how to set its parameters, and how to evaluate performance.

Please provide a more detailed abstract of your tutorial (again, see last years tutorials).

Machine learning is the task of extracting knowledge from data, often with the goal to generalize to new, unseen data. Applications of machine learning now touch nearly every aspect of everyday life, from the face detection in our

amueller

from sklearn.cluster import KMeans
from time import time
from sklearn.datasets import load_digits, fetch_mldata, load_iris, fetch_20newsgroups_vectorized

def bench_kmeans(data, n_clusters=5, init='random', n_init=1):
    start = time()
    km1 = KMeans(algorithm='lloyd', n_clusters=n_clusters, random_state=0, init=init, n_init=n_init).fit(X)
    print("lloyd time: %f inertia: %f" % (time() - start, km1.inertia_))
    start = time()
    km2 = KMeans(algorithm='elkan', n_clusters=n_clusters, random_state=0, init=init, n_init=n_init).fit(X)

amueller

amueller

amueller

from sklearn import datasets, manifold
from sklearn.neighbors import NearestNeighbors
import numpy as np


n_points = 1000
n_neighbors = 10
out_dim = 2 

n_trials = 100

amueller

import numpy as np

from sklearn import datasets
from sklearn.cross_validation import ShuffleSplit
from sklearn.grid_search import GridSearchCV
from sklearn.svm import SVC
from sklearn.preprocessing import Scaler

#data = datasets.load_digits()
data = datasets.fetch_mldata("usps")