nkthiebaut

[alias]
	st = status
	di = diff
	co = checkout
	ci = commit
	br = branch
	sta = stash
	llog = log --date=local
	flog = log --pretty=fuller --decorate
	lg = log --graph --abbrev-commit --pretty=format:'%Cred%h%Creset -%C(yellow)%d%Creset %s %Cgreen(%cr) %C(bold blue)<%an>%Creset'

nkthiebaut

git config --global credential.helper store
git config --global user.email [email protected]
git config --global user.name bobby

export PS1="\[\033[38;5;133m\]\t\[$(tput sgr0)\] \[$(tput sgr0)\]\[\033[38;5;33m\]\w\[$(tput sgr0)\] \[$(tput sgr0)\]\[\033[38;5;2m\]\\$\[$(tput sgr0)\] \[$(tput sgr0)\]"
alias ls='ls --color=auto'

nkthiebaut

# 0: Attach a lifecycle configuration to you Notebook instance to persist environments between sessions
# See https://modelpredict.com/sagemaker-save-your-conda-environments/ for instruction
# Alternative: create environments under the persisted folder (/home/ec2-user/SageMaker) with 
# conda create -p /home/ec2-user/SageMaker/py310 python=3.10
# then systematically symlink: ln -s /home/ec2-user/SageMaker/py310 ~/anaconda3/envs/py310

# 1: Create a new conda environment and make it accessible to SageMaker
conda create -y -n py310 python=3.10
conda activate py310
# From https://forum.deepchem.io/t/deploying-a-deepchem-sagemaker-notebook-instance/174

nkthiebaut

pip install -q jupyterlab
ssh-keygen -q -t rsa -N '' -f ~/.ssh/id_rsa <<<y >/dev/null 2>&1
eval "$(ssh-agent -s)" 
ssh-add
jupyter lab --ip=0.0.0.0 --port=8989 --allow-root & ssh -o StrictHostKeyChecking=no -R 80:localhost:8989 ssh.localhost.run

nkthiebaut

import tarfile
import subprocess
import os
import s3fs

REQUIRED_AWS_ENV = {"AWS_ACCESS_KEY_ID", "AWS_SECRET_ACCESS_KEY", "AWS_DEFAULT_REGION"}

def setup_aws_fs() -> s3fs.S3FileSystem:
    """Check the presence of AWS required environment variables
    and return an S3 filesystem object.

nkthiebaut

#!/usr/bin/env bash
# From https://betterdev.blog/minimal-safe-bash-script-template/

set -Eeuo pipefail
trap cleanup SIGINT SIGTERM ERR EXIT

script_dir=$(cd "$(dirname "${BASH_SOURCE[0]}")" &>/dev/null && pwd -P)

usage() {
  cat <<EOF

nkthiebaut

function onNewRow(e) {
 var sheet = SpreadsheetApp.getActiveSheet();
  Logger.log("Trigger: new row inserted:" + sheet.getLastRow());
  // Get index of row inserted
  var row = sheet.getLastRow();
  // Get word/phrase inserted
  var range = sheet.getRange(row, 1);
  var phrase = range.getValue();
  Logger.log("Row insertion detected");

nkthiebaut

from sklearn.metrics import roc_curve, auc

def plot_roc_curve(true_labels, scores):
    """ Plot ROC curve with associated score thresholds """
    # compute fpr, tpr, thresholds and roc_auc
    fpr, tpr, thresholds = roc_curve(true_labels, scores)
    roc_auc = auc(fpr, tpr) # compute area under the curve

    fig, ax = plt.subplots()

nkthiebaut

# Official doc: https://docs.python.org/3/howto/logging-cookbook.html
import os

import logging

from logging import StreamHandler
from logging import Formatter

LOGS_DIR = os.path.join(os.path.dirname(os.path.realpath(__file__)), '../logs')
if not os.path.exists(LOGS_DIR):

nkthiebaut

# inspired by https://stackoverflow.com/a/50658993/5174617
import pandas as pd
import json

df = pd.DataFrame([['0 {"a":"1","b":"2","c":"3"}'],['1 {"a" :"4","b":"5","c":"6"}']], columns=['json'])
exploded_df = df['json'].apply(json.loads).apply(pd.Series)