Skip to content

Instantly share code, notes, and snippets.

View bstadlbauer's full-sized avatar

bstadlbauer bstadlbauer

View GitHub Profile
@bstadlbauer
bstadlbauer / high_worker_usage.py
Created October 27, 2021 10:42
Example using around 100 propeller workers when started
# Copyright (C) 2015-2021 Blackshark.ai GmbH. All Rights reserved. www.blackshark.ai
import logging
import multiprocessing as mp
import uuid
from typing import Optional, List
from flytekit import map_task, workflow, task, Resources
from flytekit.interfaces.data.data_proxy import FileAccessProvider
from flytekit.loggers import logger
from flytekit.remote import FlyteRemote
import time
import geopandas as geopd
from shapely.geometry import Point, Polygon
import numpy as np
_NUMBER_OF_POINTS = int(1e6)
np.random.seed(123)
@bstadlbauer
bstadlbauer / numba_segfault.py
Created March 15, 2021 20:35
Numba jitclass segmentation fault
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
This example demonstrates jitclasses and deferred types for writing a
singly-linked-list.
"""
from __future__ import print_function, absolute_import
from collections import OrderedDict
import numpy as np
import datetime
import pandas as pd
_TIMESTAMP_FORMAT = '%Y-%m-%d %H:%M'
time_ranges = [
(datetime.datetime(2019, 6, 16, 14, 2), datetime.datetime(2019, 6, 17, 7, 2)),
(datetime.datetime(2019, 8, 15, 12, 30), datetime.datetime(2019, 8, 17, 8, 30))
]
@bstadlbauer
bstadlbauer / check_all_worker_node_environments.py
Created September 26, 2019 08:41
Check python environment configuration on all spark worker nodes
import socket
import pyspark
from pip._internal.operations import freeze
NUMBER_OF_OPERATIONS = 100_000
NUMBER_OF_PARTITIONS = 10_000
def get_environment_and_hostname(_):
hostname = socket.gethostname()