Skip to content

Instantly share code, notes, and snippets.

View romanegloo's full-sized avatar
😀
clean-up

Jiho Noh romanegloo

😀
clean-up
  • University of Kentucky
  • Lexington, KY
View GitHub Profile
@romanegloo
romanegloo / README.md
Last active April 20, 2021 02:35
GitHub README template

[title]

project description

Requirements

list of required packages or softwares, versions of systems

Data

@romanegloo
romanegloo / fib.md
Last active October 24, 2019 00:36

(Note. If you missed the "Python Primer" lab and you have no Python experience, please come to Jiho's office hours. We can do this together.)

Fibonacci numbers are defined recursively as below:

f(0) = 0, f(1) = 1
f(n) = f(n-1) + f(n-2)

Implement a fibonacci function fib(n), and use it to print out the Fibonacci numbers where n ranges from 0 to 9 such as:

@romanegloo
romanegloo / uts_api_client.py
Created September 13, 2019 14:46
UMLS API client
""" UTS (UMLS Terminology Services) API client """
import json
from pathlib import Path
import requests
from lxml.html import fromstring
class UtsClient:
"""All the UTS REST API requests are handled through this client"""
def __init__(self, apikey=None):
@romanegloo
romanegloo / cui2mesh.py
Created September 6, 2019 15:01
Find corresponding MeSH terms from CUIs in evaluation datasets
#!/usr/bin/env python3
# pylint: disable=invalid-name
"""Reads UMNSRS datasets where CUIs are used, add corresponding MeSHes to the
CUIs"""
from pathlib import Path
import csv
from tqdm import tqdm
from BMET.uts_api_client import UtsClient
@romanegloo
romanegloo / build_ES_pubtator_index.py
Created September 5, 2019 19:31
build_ES_pubtator_index: Building ElasticSearch Index of PubTator
#!/usr/bin/env python3
"""
Parse PubTator raw data file, and creates ElasticSearch index, and populate the
corpus (index: 'pubtator')
You can download PubTator dataset from the following link:
ftp://ftp.ncbi.nlm.nih.gov/pub/lu/PubTator/bioconcepts2pubtator_offsets.gz
Place the file in 'data/' and run this code. Don't need to decompress the file.
We assume that ElasticSearch is properly installed and accessible via its API
@romanegloo
romanegloo / script_gen_pubtator_db.py
Created August 30, 2019 13:58
A script that reads MeSH descriptors and PubTator doc data from data files and create SQLite database to store the encoded docs for later training uses. (deprecated)
#!/usr/bin/env python3
"""Preprocess PubTator corpus and ScopeNotes of MeSH descriptors for language
model training (LmBMET).
1) Given the original PubTator biocepts annotated documents, this interpolates
the concept codes into document texts. Before that, this will count word
frequencies and generate vocabulary which will include the entire set of
bioconcepts (MeSH in particular). In case that a pre-trained embeddings file
(.vec) is provided, we obtain a vocabulary from the embeddings.
import csv
import sqlite3
eval_file = "data/eval/MayoSRS_mesh.csv"
db_file = "data/pubtator/pubtator-20190725-6496be10.db"
words = []
with open(eval_file) as f:
csv_reader = csv.DictReader(f)
for row in csv_reader:
@romanegloo
romanegloo / ukycs215_lab8.md
Last active March 29, 2019 14:23
lab8 note

Lab 08

instruction note (lab8)

Lab 9 is canceled (lab 8 is extended by a week)

Reviews

  1. include contact.h from contactList.h. Use double quotation marks
  2. #pragma once tells the compiler to include the source code only once
@romanegloo
romanegloo / ukycs215_lab7.md
Last active March 7, 2019 16:43
lab7 note
@romanegloo
romanegloo / ukycs215_lab5.md
Created February 13, 2019 20:27
lab05_note

Lab 05

Lab 05 instruction

Easiest so far!!!

Checklist

  • menu option is single char
  • use do-while for the menu loop