ksindi’s gists

ksindi / k8s-svc-annotations.md

Created December 8, 2017 18:14 — forked from mgoodness/k8s-svc-annotations.md

AWS ELB-related annotations for Kubernetes Services (v1.5)

AWS Service annotations

service.beta.kubernetes.io/aws-load-balancer-access-log-emit-interval (in minutes)
service.beta.kubernetes.io/aws-load-balancer-access-log-enabled (true|false)
service.beta.kubernetes.io/aws-load-balancer-access-log-s3-bucket-name
service.beta.kubernetes.io/aws-load-balancer-access-log-s3-bucket-prefix
service.beta.kubernetes.io/aws-load-balancer-backend-protocol (http|https|ssl|tcp)
service.beta.kubernetes.io/aws-load-balancer-connection-draining-enabled (true|false)
service.beta.kubernetes.io/aws-load-balancer-connection-draining-timeout (in seconds)

ksindi / pagerank.py

Created December 7, 2017 15:00 — forked from diogojc/pagerank.py

python implementation of pagerank

	import numpy as np
	from scipy.sparse import csc_matrix

	def pageRank(G, s = .85, maxerr = .001):
	"""
	Computes the pagerank for each of the n states.

	Used in webpage ranking and text summarization using unweighted
	or weighted transitions respectively.

ksindi / serialize-numpy-array.py

Created November 20, 2017 13:16 — forked from alexland/serialize-numpy-array.py

serialize, persist, retrieve, and de-serialize a NumPy array as a binary string (any dimension, any dtype); exemplary use case: a web app calculates some result--eg, from a Machine Learning algorithm, using NumPy and the result is a NumPy array; it is efficient to just return that result to rather than persist the array then retrieve it via query


	import time
	import numpy as NP
	from redis import StrictRedis as redis

	# a 2D array to serialize
	A = 10 * NP.random.randn(10000).reshape(1000, 10)

	# flatten the 2D NumPy array and save it as a binary string
	array_dtype = str(A.dtype)

ksindi / pdio.py

Created November 19, 2017 14:42 — forked from luispedro/pdio.py

Save & load from a pandas DataFrame/Series

	import numpy.lib
	import numpy as np
	import pandas as pd
	import cPickle as pickle

	def save_pandas(fname, data):
	'''Save DataFrame or Series

	Parameters
	----------

ksindi / fast_cosine_matrix.py

Last active December 14, 2017 01:39

	"""

	Notes:
	- Is about 2ms for (100, (10000, 100)) shape inputs on my i7 laptop
	- It's 2x faster without doing vector normalize (might make sense to pre-normalize the vectors)
	"""

	import numpy as np
	import numba

ksindi / gist:09a0fb58d479b55d6168d9094ce38d42

Created November 14, 2017 15:38 — forked from FedericoV/gist:0e7d6d8c8794a99a7a42

Cosine Similarity that handles NaN with Numba

	import numba

	@numba.jit(target='cpu', nopython=True)
	def fast_cosine(u, v):
	m = u.shape[0]
	udotv = 0
	u_norm = 0
	v_norm = 0
	for i in range(m):
	if (np.isnan(u[i])) or (np.isnan(v[i])):

ksindi / inception_annoy.py

Created November 13, 2017 16:27 — forked from thomasdullien/inception_annoy.py

Inception for feature extraction, ANNoy for nearest-neighbor search

	"""
	Simple, hacked-up image similarity search using Tensorflow + the inception
	CNN as feature extractor and ANNoy for nearest neighbor search.

	Requires Tensorflow and ANNoy.

	Based on gist code under
	https://gist.github.com/david90/e98e1c41a0ebc580e5a9ce25ff6a972d
	"""
	from annoy import AnnoyIndex

ksindi / pdf_to_image.py

Created August 28, 2017 15:02

	"""

	Dockerfile

	FROM python:3.6

	RUN apt-get install -y ghostscript \
	libmagickwand-dev

	RUN pip install wand PyPDF2

ksindi / pyarrow-segfault.py

Last active August 20, 2017 19:42

	import os

	import pandas as pd
	import pyarrow as pa
	import pyarrow.parquet as pq

	NUM_FIELDS = 600
	NUM_TABLES = 100

	fields = [

ksindi / pyarrow-segfault.py

Created August 20, 2017 18:42

	import pandas as pd
	import pyarrow as pa
	import pyarrow.parquet as pq

	fields = [
	pa.field('column1', pa.string()),
	pa.field('column2', pa.int64()),
	pa.field('column3', pa.string()),
	]

Kamil Sindi ksindi

AWS Service annotations