jnothman

from sklearn.feature_extraction.text import CountVectorizer
from joblib import Memory
from sklearn.base import clone
from sklearn.datasets import fetch_20newsgroups


class CachedTransformMixin:
    memory = Memory('/tmp/cache')

    def transform(self, *args, **kwargs):

jnothman

SELECT
?subj
?subjLabel
?prop
?position
?positionLabel
?start
?end
?district
?districtLabel

jnothman

import numpy as np
from sklearn import datasets
from sklearn.semi_supervised import LabelPropagation, LabelSpreading
###for n_samples in [20, 200, 2000, 20000]:
###    X, y = datasets.make_classification(n_samples=n_samples, n_classes=3, n_informative=3)
for (X, y) in [datasets.load_iris(return_X_y=True)]:
    for model in [LabelPropagation(max_iter=1000),
                  #LabelSpreading(alpha=0.01),
                  #LabelSpreading(alpha=0.1),
                  #LabelSpreading(alpha=0.3)

jnothman

from doc2dash.parsers.intersphinx import (InterSphinxParser,
                                          inv_entry_to_path,
                                          ParserEntry)
import doc2dash.parsers

class InterSphinxWithUserGuide(InterSphinxParser):
    def convert_type(self, inv_type):
        if inv_type == 'std:doc':  # sphinx type
            return 'Guide'  # Dash type
        return super(InterSphinxWithUserGuide, self).convert_type(inv_type)

jnothman

package:
  name: cssdecl
  version: "0.1.1"

source:
  fn: cssdecl-0.1.1.tar.gz
  url: https://pypi.python.org/packages/c8/6a/5620e9f501f2332fe11fa3fc227a73458dc0d0ac43fce81d622906708789/cssdecl-0.1.1.tar.gz
  md5: b6f421becf8f14843de7bf821ccd80c2
#  patches:
   # List any patch files here

jnothman

#!/usr/bin/osascript

on run argv
        set l to 0
        set t to 0
        repeat with j from 1 to (count argv)
                set a to item j of argv
                tell application a
                        repeat with x from 1 to (count windows)
                                set b to bounds of window x

jnothman

# coding: utf-8

import pandas as pd
import numpy as np


df = pd.DataFrame(np.random.rand(5, 10))

jnothman

from __future__ import print_function
from collections import Counter

import nltk


def count_ngrams(tokens, min_unigram_freq=2, min_ngram_freq=5, max_n=5):
    n_tokens = len(tokens)
    print('Number of tokens:', n_tokens)
    unigram_freqs = Counter(tokens)

jnothman

"""
============================
Classifier Chain
============================
An ensemble of 10 logistic regression classifier chains trained on a
multi-label dataset achieves a higher Jaccard similarity score than a set
of independently trained logistic regression models.

"""

jnothman

import numpy as np

class Resample(object):
    def __init__(self, cv, method='under'):
        self.cv = cv
        self.method = method

    def split(self, X, y, **kwargs):
        for train_idx, test_idx in self.cv.split(X, y, **kwargs):
            counts = np.bincount(y[train_idx])  # assumes y are from {0, 1..., n_classes-1}