undarmaa

undarmaa

#!/usr/bin/env python
# -*- encoding: utf-8 -*-
# based on http://scikit-learn.org/stable/auto_examples/document_clustering.html

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.cluster import KMeans, MiniBatchKMeans
from sklearn.metrics.pairwise import pairwise_distances
import numpy as np
from time import time
from collections import defaultdict

undarmaa

텐서플로우 시작하기

글쓴이: 김정주(haje01@gmail.com)

이 문서는 텐서플로우 공식 페이지 내용을 바탕으로 만들어졌습니다.

---

소개

텐서플로우(TensorFlow)는 기계 학습과 딥러닝을 위해 구글에서 만든 오픈소스 라이브러리입니다. 데이터 플로우 그래프(Data Flow Graph) 방식을 사용하였습니다.

undarmaa

from sklearn.metrics.pairwise import pairwise_distances
import numpy as np

# X shoudl be a numpy matrix, very likely sparse matrix: http://docs.scipy.org/doc/scipy-0.14.0/reference/generated/scipy.sparse.csr_matrix.html#scipy.sparse.csr_matrix
# T1 > T2 for overlapping clusters
# T1 = Distance to centroid point to not include in other clusters
# T2 = Distance to centroid point to include in cluster
# T1 > T2 for overlapping clusters
# T1 < T2 will have points which reside in no clusters
# T1 == T2 will cause all points to reside in mutually exclusive clusters

undarmaa

# iPython Notebook with per-user storage and config
# 
# Based on crosbymichael/ipython
# Creates a Docker image with IPython Notebook installed.
#
# It expects to be run like this: 
#
# docker run -v /home/eduard/notebooks/eduard:/notebooks benthoo/ipython-user
#
# You provide a folder per user on the host system. This folder will hold the users notebooks and also needs to contain the

undarmaa

from sklearn.datasets.samples_generator import make_regression
from sklearn.linear_model import Lasso
import numpy as np
 
X, y = make_regression(n_samples=200, n_features=5000, random_state=0)
 
alpha = 1
model = Lasso(alpha=alpha, fit_intercept=False, max_iter=1000)
 
model.fit(X, y)

undarmaa

import pandas as pd
import pylab as pl
from sklearn.neighbors import KNeighborsClassifier


df = pd.read_csv("https://s3.amazonaws.com/demo-datasets/wine.csv")

test_idx = np.random.uniform(0, 1, len(df)) <= 0.3
train = df[test_idx==True]
test = df[test_idx==False]

undarmaa

import numpy as np
import numpy
import theano
import theano.tensor as T
from theano import function, config, shared, sandbox
from theano import ProfileMode
from sklearn import cluster, datasets
import matplotlib.pyplot as plt

def rsom(data, cluster_num, alpha, epochs = -1, batch = 1, verbose = False):