🍉

Anja Pilz aplz

🍉

Machine Learning, Natural Language Processing and Information Retrieval for Medicine & Healthcare.

aplz / test_parameterized_exception.py

Last active August 5, 2021 09:48

assert that some input results in an exception being raised

	import pytest
	from typing import List

	def method_under_test(values: List[int]) -> int:
	"""dummy method"""
	if 0 in values:
	raise(ValueError("Zeros are not supported"))
	return sum(values)

	@pytest.mark.parametrize(

aplz / numpy_vector_to_dlib_sparse_vector.py

Last active November 26, 2020 08:21

Create a sparse vector for dlib from a numpy array

	"""How to create a sparse vector for dlib from a numpy array.

	dlib: http://dlib.net/
	"""
	import dlib
	import numpy

	vector = [0,2,0]
	sv = dlib.sparse_vector()
	for i in numpy.nonzero(vector)[0]:

aplz / analyze_ngram.js

Created July 10, 2020 12:04

elasticsearch analyze API with customized ngram analyzer

aplz / gist:f4eea39b5a5e80ca28cfeb71872f77fd

Created March 23, 2020 11:13 — forked from joelkuiper/gist:4869d148333f279c2b2e

All UMLS semantic types

aplz / scroll.py

Created October 16, 2019 09:16 — forked from hmldd/scroll.py

Example of Elasticsearch scrolling using Python client

	# coding:utf-8

	from elasticsearch import Elasticsearch
	import json

	# Define config
	host = "127.0.0.1"
	port = 9200
	timeout = 1000
	index = "index"

aplz / disable_spell_checker.txt

Last active April 17, 2019 08:45

locally disable spell checking in pycharm and IDEA

	place the line below above the offensive code part
	# noinspection SpellCheckingInspection

	for IDEA, the analogon is
	@SuppressWarnings("SpellCheckingInspection")

aplz / es_german_analyzer.py

Last active October 16, 2019 09:17

elasticsearch-py german analyzer / tokenizer

	es = Elasticsearch()
	tokens = es.indices.analyze(
	body={"analyzer": "german",
	"text": "Die junge Informatikerin Katie Bouman machte die "
	"historische Aufnahme eines schwarzen Lochs "
	"möglich."})['tokens']
	for token in tokens:
	print(token)

aplz / bulk_upsert_generator.py

Created March 25, 2019 15:12

elasticsearch - generator for bulk ingestion with upserts

	for doc in collection:
	yield {
	'_id': id, '_type': TYPE,
	'_index': index,
	# with "doc_as_upsert": True, the document is either updated
	# or created if it does not already exist.
	'_op_type': 'update', 'doc': doc, "doc_as_upsert": True
	}

aplz / copy_directory_filtered_by_timestamp_range.sh

Last active October 24, 2018 12:15

filter files in directory by creation date, copy them to a new directory preserving original sub-directory structure

find . -type f -newermt '2018/10/14' -not -newermt '2018/10/21'| cpio -pvdmB <your-directory>

aplz / copy_files_by_timestamp_range.sh

Last active October 24, 2018 12:02

copy files by creation timestamp range in shell

for i in `find . -type f -newermt '2018/10/14' -not -newermt '2018/10/21'`; do cp -p $i <your_folder>/; done