darkseed

VGG_ILSVRC_19_layers_train_val.prototxt
name: "VGG_ILSVRC_19_layers"
layers {
  name: "data"
  type: DATA
  include {
    phase: TRAIN
  }
 transform_param {
    crop_size: 224

darkseed

from mincepie import mapreducer, launcher
import gflags
import glob
import leargist
import numpy as np
import os
from PIL import Image
import uuid
 
# constant value

darkseed

class BloomFilter(val size: Int, val expectedElements: Int){
  require(size > 0)
  require(expectedElements > 0)

  val bitArray = new BitArray(size)
  val k = Math.ceil((bitArray.size / expectedElements) * Math.log(2.0)).toInt
  val expectedFalsePositiveProbability = Math.pow(1 - Math.exp(-k * 1.0 * expectedElements / bitArray.size), k)

  def add(hash: Int) {
    def add(i: Int, seed: Int) {

darkseed

/*
Copyright 2012 Twitter, Inc.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and

darkseed

import breeze.linalg._
import breeze.stats._
import scala.math.sqrt


/**
 * Effecient for sparse vectors. Scales in O(activeSize)
 */
// Must take SparseVector, for implicits to be linked correctly
def pearson(a: SparseVector[Double], b: SparseVector[Double]): Double = {

darkseed

// genrate all combinations of integers in range from 0 to `len`-1
// fast as fuck
def combIdxs(len: Int, k: Int): Iterator[Array[Int]] = {
  val arr = Array.range(0, k)
  arr(k-1) -= 1
  val end = k-1
  
  Iterator.continually {
    arr(end) += 1
    if (arr(end) >= len) {

darkseed

// min-hash

val fs: Vector[Int => Int] // hash funkce

items map { it => fs map { f => f(it) } } fold (vectorPairwise(min), initialValue = Vector.fill(infinity))



// HyperLogLog

darkseed

import breeze._
import breeze.linalg._
import breeze.numerics._
import java.awt.image.BufferedImage
import javax.imageio.ImageIO

val f = ???
val img = javax.imageio.ImageIO.read(new File(f))
val gray = new BufferedImage(img.getWidth, img.getHeight, BufferedImage.TYPE_BYTE_GRAY)
val g = gray.createGraphics()

darkseed

import breeze.linalg._
import breeze.stats
import breeze.numerics._

val dataFile = new File(???)
val userItems: Array[SparseVector[Double]] = loaderUserItemsWithRatings(dataFile, """[ ,:]""".r)
val itemUsers: Array[SparseVector[Double]] = transpose(userItems) map { vec => normalize(vec, 2) }

// weights
val N = DenseVector.fill[Double](itemIndex.size)(userIndex.size) // vector where total numbers of users is repeated

darkseed

package breeze.linalg.functions

import breeze.generic.UFunc
import breeze.linalg.{SparseVector, DenseVector}
import breeze.numerics.sqrt

/**
 * A weighted Euclidean distance function implementation
 */
object weightedEuclideanDistance extends UFunc {