ellimilial

## string_to_deterministic_uniform_float.py
import hashlib

def string_to_deterministic_float(s: str) -> float:
    """
    Given a string, deterministically map it to a uniformly sampled float in range [0.0, 1.0).
    """
    b = bytes(s, encoding='utf-8')
    h_dig = hashlib.sha256(b).hexdigest()
    return int(h_dig, base=16) / 16 ** len(h_dig)

## string_to_deterministic_float.py
import hashlib


def string_to_deterministic_float(s: str) -> float:
    """
    Given a string, deterministically convert it to a range [0.0, 1.0).
    """
    b = bytes(s, encoding='utf-8')
    h_dig = hashlib.sha256(b).hexdigest()
    return int(h_dig, base=16) / 16 ** len(h_dig)

## blog_pl_scraper.py
from collections import OrderedDict

import os
import bs4
import requests
import re
import json

from requests.adapters import HTTPAdapter
from urllib3 import Retry

## hadoop_benchmark_terasort_multiple_runs.sh
#!/bin/bash

# Script to run hadoop terasort benchmark specified amount of times, getting the average runtime for all 3 stages.
#
# To run via ssh, say on Jenkins, wrap in:
# ssh namenode.server.com <<'ENDSSH'
# (... code ...)
# ENDSSH

readonly EXAMPLES_JAR="(...)/hadoop-mapreduce-examples.jar"

## hadoop_benchmark_DFSIO_read_write_multiple_runs.sh
#!/bin/bash

#
# Run DFSIO write and read tests for multiple file size/count configurations. Get the average speed over RUNS_PER_CONFIG executions.
# The throughtput calculation method assumes all tests are run on a single 'wave' , i.e. BATCH_SIZE < total mapper task
#
# Replace hadoop/yarn in run commands as required.
#
# To run via ssh, say on Jenkins, wrap in:
# ssh namenode.server.com <<'ENDSSH'

## gist:5ef1d1917e00970d4457
sudo apt-get install python-dev cython git python-pip
sudo pip install git+git://github.com/ellimilial/gevent.git@master
	import hashlib

	def string_to_deterministic_float(s: str) -> float:
	"""
	Given a string, deterministically map it to a uniformly sampled float in range [0.0, 1.0).
	"""
	b = bytes(s, encoding='utf-8')
	h_dig = hashlib.sha256(b).hexdigest()
	return int(h_dig, base=16) / 16 ** len(h_dig)
	import hashlib


	def string_to_deterministic_float(s: str) -> float:
	"""
	Given a string, deterministically convert it to a range [0.0, 1.0).
	"""
	b = bytes(s, encoding='utf-8')
	h_dig = hashlib.sha256(b).hexdigest()
	return int(h_dig, base=16) / 16 ** len(h_dig)
	from collections import OrderedDict

	import os
	import bs4
	import requests
	import re
	import json

	from requests.adapters import HTTPAdapter
	from urllib3 import Retry
	#!/bin/bash

	# Script to run hadoop terasort benchmark specified amount of times, getting the average runtime for all 3 stages.
	#
	# To run via ssh, say on Jenkins, wrap in:
	# ssh namenode.server.com <<'ENDSSH'
	# (... code ...)
	# ENDSSH

	readonly EXAMPLES_JAR="(...)/hadoop-mapreduce-examples.jar"
	#!/bin/bash

	#
	# Run DFSIO write and read tests for multiple file size/count configurations. Get the average speed over RUNS_PER_CONFIG executions.
	# The throughtput calculation method assumes all tests are run on a single 'wave' , i.e. BATCH_SIZE < total mapper task
	#
	# Replace hadoop/yarn in run commands as required.
	#
	# To run via ssh, say on Jenkins, wrap in:
	# ssh namenode.server.com <<'ENDSSH'
	sudo apt-get install python-dev cython git python-pip
	sudo pip install git+git://github.com/ellimilial/gevent.git@master